OK, one more try I thought.....
I took a fresh 3.10.2-7179 image, set my preferences, opened up the Packages Universe, carefully selected the packages I wanted, being extra careful to avoid the ones that I've had problems with already, and started the installs......
Things looked like they had finished fine, and so I tried clicking on the Packages Universe window again to see what it might say and KABOOM! (this all retyped by hand)
***System error handling failed*** Original error: MessageNotUnderstood: LabelMorph class>>contents:. Debugger error: MessageNotUnderstood: LabelMorph class>>contensts:: [] in Debugger class>>openOn:context:label:contents:fullView: {[:ex | self primitiveError: 'Original error: ', title asString, ', Debuge...]} BlockContext>>valueWithPossibleArgs: [] in MethodContext(ContextPart)>>handleSignal: {[(self tempAt: 2) valueWithPossibleArgs: {exception}]} blockContext>>ensure: MethodContext(ContextPart)>>handleSignal: MessageNotUnderstood(Exception)>>signal LabelMorph class(Object)>>doesNotUnderstand: #contents: UIThemeSoftSqueak(UITheme)>>buttonLabelForText: UIThemeSoftSqueak>>buttonLabelForText: UIThemeSoftSqueak(UITheme)>>buttonLabelFor: PluggableButtonMorph>>newLabel PluggableButtonMorph>>label: [] in Debugger>>buttonRowForPreDebugWindow: {[:quad | aButton := PluggableButtonMorph on: aDebugWindow getState:...]} OrderedCollection>>do: Debugger>>buttonRowForPreDebugWindow: Debugger>>buildMorphicNotifierLabelled:message: Debugger>>openNotifierContents:label: [] in Debugger class>>openOn:context:label:contents:fullView: {[debugger := self new process: process controller: controller co...]} BlockContext>>on:do: [] in Debugger class>>openOn:context:label:contents:fullView: {[[debugger := self new process: process controller: controller co...]} --------------------------- Type CR to enter an emergency evaluator. Type any other character to restart.
Sadly it doesn't look like there's even any obvious hint as to what package might have caused the problem.
I'm beginning to think this is a hopeless cause. Clearly not even a majority of packages in the currently published universe for 3.10.2 have ever actually been tested.
After restarting the VM (I hit CR, then typed "exit<CR>" to the evaluator) things seemed to start working again so I've copied the text out of the Universe window's left-hand pane to show what things I've installed (and so far as I can see everything is now marked as "(installed)").
As I said, things _seemed_ to start working again. Most things actually don't work -- I can't open any kind of browser, for example -- I just get another very similar crash.
Sigh.
If anyone can see anything out of the "(installed)" packages that I shouldn't have installed for any reason, please let me know and I'll try again without whatever and see what happens.
Code Browsing AutomaticMethodCategorizer version 0.24 AutomaticMethodCategorizer version 0.25 AutomaticMethodCategorizerOB version 0.1 AutomaticMethodCategorizerOB version 0.2 AutomaticMethodCategorizerServices version 0.2 DependencyBrowser version 0.1 OB-Enhancements version 0.61 OB-Enhancements version 0.101 OB-Enhancements version 0.115 OB-Enhancements version 0.138 OB-Enhancements version 0.145 OB-Enhancements version 0.171 OB-Enhancements version 0.174 OB-Enhancements version 0.176 OB-Enhancements version 0.191 OB-Enhancements version 0.203 OB-Enhancements version 0.219 OB-Enhancements version 0.220 OB-Enhancements version 0.221 OB-Enhancements version 0.247 OB-Enhancements version 0.260 (installed) OB-Enhancements version 0.265 OBProtocolBrowser version 0.16 OBProtocolBrowser version 0.17 OBProtocolBrowser version 0.18 OBProtocolBrowser version 0.19 OmniBrowser version 0.323 OmniBrowser version 0.335 OmniBrowser version 0.337 OmniBrowser version 0.341 OmniBrowser version 0.342 OmniBrowser version 0.347 OmniBrowser version 0.350 OmniBrowser version 0.351 OmniBrowser version 0.357 OmniBrowser version 0.358 OmniBrowser version 0.359 OmniBrowser version 0.371 OmniBrowser version 0.376 OmniBrowser version 0.378 OmniBrowser version 0.380 OmniBrowser version 0.381 OmniBrowser version 0.382 OmniBrowser version 0.387 OmniBrowser version 0.388 OmniBrowser version 0.391 OmniBrowser version 0.392 OmniBrowser version 0.393 OmniBrowser version 0.404 OmniBrowser version 0.405 OmniBrowser version 0.406 OmniBrowser version 0.408 OmniBrowser version 0.412 OmniBrowser version 0.412.1 OmniBrowser version 0.413 OmniBrowser version 0.416 OmniBrowser version 0.419 OmniBrowser version 0.420 OmniBrowser version 0.426 (installed) OmniBrowser version 0.432 OmniBrowser-Full version 0.1 OmniBrowser-Full version 0.2 OmniBrowser-Full version 0.3 OmniBrowser-Full version 0.4 OmniBrowser-Full version 0.7 OmniBrowser-Full version 0.8 OmniBrowser-Full version 0.9 OmniBrowser-Full version 0.10 OmniBrowser-Full version 0.14 OmniBrowser-Full version 0.18 OmniBrowser-Full version 0.19 OmniBrowser-Full version 0.20 OmniBrowser-Full version 0.24 OmniBrowser-Full version 0.25 OmniBrowser-Full version 0.26 (installed) OmniBrowser-Full version 0.27 OmniBrowser-Morphic version 0.2 OmniBrowser-Morphic version 0.3 OmniBrowser-Morphic version 0.4 OmniBrowser-Morphic version 0.5 OmniBrowser-Morphic version 0.6 OmniBrowser-Morphic version 0.8 OmniBrowser-Morphic version 0.9 OmniBrowser-Morphic version 0.10 OmniBrowser-Morphic version 0.13 OmniBrowser-Morphic version 0.14 OmniBrowser-Morphic version 0.18 OmniBrowser-Morphic version 0.18.2 OmniBrowser-Morphic version 0.20 OmniBrowser-Morphic version 0.21 OmniBrowser-Morphic version 0.22 OmniBrowser-Morphic version 0.23 OmniBrowser-Morphic version 0.24 OmniBrowser-Morphic version 0.27 OmniBrowser-Morphic version 0.28 OmniBrowser-Morphic version 0.33 OmniBrowser-Morphic version 0.34 OmniBrowser-Morphic version 0.38 OmniBrowser-Morphic version 0.39 OmniBrowser-Morphic version 0.40 OmniBrowser-Morphic version 0.43 OmniBrowser-Morphic version 0.45 OmniBrowser-Morphic version 0.46 OmniBrowser-Morphic version 0.47 OmniBrowser-Morphic version 0.48 OmniBrowser-Morphic version 0.49 OmniBrowser-Morphic version 0.50 OmniBrowser-Morphic version 0.51 OmniBrowser-Morphic version 0.52 OmniBrowser-Morphic version 0.52.1 OmniBrowser-Morphic version 0.58 OmniBrowser-Morphic version 0.58.1 OmniBrowser-Morphic version 0.59 OmniBrowser-Morphic version 0.62 (installed) OmniBrowser-Morphic version 0.63 OmniBrowser-SUnitIntegration version 0.1 OmniBrowser-SUnitIntegration version 0.2 OmniBrowser-SUnitIntegration version 0.4 OmniBrowser-SUnitIntegration version 0.5 OmniBrowser-SUnitIntegration version 0.8 OmniBrowser-SUnitIntegration version 0.9 OmniBrowser-SUnitIntegration version 0.9.1 OmniBrowser-SUnitIntegration version 1 obsolete OmniBrowser-Standard version 0.166 OmniBrowser-Standard version 0.182 OmniBrowser-Standard version 0.183 OmniBrowser-Standard version 0.186 OmniBrowser-Standard version 0.188 OmniBrowser-Standard version 0.192 OmniBrowser-Standard version 0.194 OmniBrowser-Standard version 0.195 OmniBrowser-Standard version 0.199 OmniBrowser-Standard version 0.201 OmniBrowser-Standard version 0.202 OmniBrowser-Standard version 0.205 OmniBrowser-Standard version 0.208 OmniBrowser-Standard version 0.211 OmniBrowser-Standard version 0.222 OmniBrowser-Standard version 0.222.2 OmniBrowser-Standard version 0.239 OmniBrowser-Standard version 0.259 OmniBrowser-Standard version 0.261 OmniBrowser-Standard version 0.265 OmniBrowser-Standard version 0.268 OmniBrowser-Standard version 0.271 OmniBrowser-Standard version 0.274 OmniBrowser-Standard version 0.282 OmniBrowser-Standard version 0.303 OmniBrowser-Standard version 0.306 OmniBrowser-Standard version 0.307 OmniBrowser-Standard version 0.310 OmniBrowser-Standard version 0.312 OmniBrowser-Standard version 0.318 OmniBrowser-Standard version 0.320 OmniBrowser-Standard version 0.323 OmniBrowser-Standard version 0.323.1 OmniBrowser-Standard version 0.325 OmniBrowser-Standard version 0.327 OmniBrowser-Standard version 0.335 OmniBrowser-Standard version 0.337 OmniBrowser-Standard version 0.345 OmniBrowser-Standard version 0.346 OmniBrowser-Standard version 0.347 OmniBrowser-Standard version 0.356 OmniBrowser-Standard version 0.369 (installed) OmniBrowser-Standard version 0.377 OmniBrowser-TraitsIntegration version 0.1 OmniBrowser-TraitsIntegration version 0.28 OmniBrowser-TraitsIntegration version 0.28.1 OmniBrowser-TraitsIntegration version 0.32 OmniBrowser-TraitsIntegration version 0.33 OmniBrowser-TraitsIntegration version 0.34 OmniBrowser-TraitsIntegration version 1 obsolete Shout version 3.15-tween.65 Shout version 3.15-tween.69 Shout version 3.15-tween.70 Shout version 3.15-tween.71 Shout version 3.15-tween.72 ShoutMonticello version 1-tween.2 ShoutOmniBrowser version tween.3 ShoutOmniBrowser version tween.4 ShoutOmniBrowser version tween.4.1 ShoutOmniBrowser version tween.5 ShoutOmniBrowser version tween.6 TraitsOmniBrowser version 0.39 VersionsBrowser version 0.3.1 VersionsBrowser version 1.0 Whisker Browser version 1.0 eCompletion version 0.83 eCompletion version 0.83.1 eCompletion version 0.87 eCompletion version 0.88 eCompletion version 0.89 eCompletion version 0.90 eCompletion version 0.91 eCompletion version 0.92 eCompletion version 0.93 eCompletion-Traits version 0.1 eCompletionOmniBrowser version 0.2 eCompletionOmniBrowser version 0.4 eCompletionOmniBrowser version 0.5 Constraints Cassowary version 0.60 (installed) Connectors version 2.3-187 Data Structures (installed) Array2D version 1 (installed) BitArray version 1.0 (installed) FixedDecimal version 13 Iterator version 1.0 KGraph version 0.0 PropertyList version 1.1 PropertyList version 5 Development DebugReport version 1.1 DebugReport version 2.5 DependencyWalker version 0.1 DictionaryBrowser version 0.2 DynamicProtocols version 0.44 DynamicProtocols version 0.45 DynamicProtocols version 0.47 DynamicProtocols version 0.48 DynamicProtocols version 0.49 DynamicProtocols version 0.50 DynamicProtocols version 0.53 DynamicProtocols version 0.56 DynamicProtocols version 0.58 DynamicProtocols version 0.59 DynamicProtocols version 0.60 DynamicProtocols version 0.67 LogEngine version 2 Logging version 3.10 Logging version 3.11 Logging version 3.27 Magritte-Model version 1.0.17.331 Magritte-Mse version 1.0.17.2 Magritte-Tests version 1.0.17.148 Mse version 1.0.17.10 NewCompiler version 157 NewCompiler version 158 NewCompiler version 162 NewCompiler version 164 NewCompiler version 172 NewCompiler version 182 NewCompiler version 184 NewCompiler version 185 NewCompiler version 187 NewCompiler version 195 NewCompiler version 202 NewCompiler version 204 NewCompiler version 215 NewCompiler version 228 NewCompiler version 231 NewCompiler version 241 NewCompiler version 247 NewCompiler version 248 NewCompiler version 251 NewCompiler version 268 NewCompiler version 270 NewCompiler version 272 NewCompiler version 277 NewCompiler version 278 NewCompiler version 280 NewCompiler version 281 NewParser version 2 NewParser version 3 NewParser version 3.1 NewParser version 4 NewParser version 5 ObjectFinder version 0.1 ObjectFinder version 0.2 OmniBrowser-Refactory version 0.1 OmniBrowser-Refactory version 0.5 OmniBrowser-Refactory version 0.7 OmniBrowser-Refactory version 0.10 OmniBrowser-Refactory version 0.16 OmniBrowser-Refactory version 0.19 OmniBrowser-Refactory version 0.20 OmniBrowser-Refactory version 0.22 OmniBrowser-Refactory version 0.23 OmniBrowser-Refactory version 0.24 OmniBrowser-Refactory version 0.26 OmniBrowser-Refactory version 0.27 OmniBrowser-Refactory version 0.30 OmniBrowser-Refactory version 0.33 OmniBrowser-Refactory version 0.33new OmniBrowser-Refactory version 0.38 OmniBrowser-Refactory version 0.41 OmniBrowser-Refactory version 0.42 OmniBrowser-Refactory version 0.44 OmniBrowser-Refactory version 0.48 OmniBrowser-Refactory version 0.53 OmniBrowser-Refactory version 0.65 OmniBrowser-Refactory version 0.69 OmniBrowser-Refactory version 0.78 OmniBrowser-Refactory version 0.79 OmniBrowser-Refactory version 0.84 OmniBrowser-Refactory version 0.85 OmniBrowser-Refactory version 0.89 OmniBrowser-Refactory version 0.94 OmniBrowser-Refactory version 0.98 OmniBrowser-Refactory version 1.0 OmniBrowser-Refactory version 101 OmniBrowser-Refactory version 101.1 OmniBrowser-Refactory version 103 OmniBrowser-Refactory version 107 OmniBrowser-Refactory version 108 (installed) OmniBrowser-Refactory version 113 OmniBrowser-Regex version 0.8 OmniBrowser-Regex version 0.8.1 OmniBrowser-Regex version 0.8.2 OmniBrowser-Regex version 0.9 (installed) OmniBrowser-Regex version 0.9.1 OmniBrowser-Tools version 0.7 OmniBrowser-Tools version 0.11 OmniBrowser-Tools version 0.12 OmniBrowser-Tools version 0.13 OmniBrowser-Tools version 0.20 OmniBrowser-Tools version 0.30 OmniBrowser-Tools version 0.31 OmniBrowser-Tools version 0.32 OmniBrowser-Tools version 0.33 OmniBrowser-Tools version 0.34 OmniBrowser-Tools version 0.35 Pier-Model version 1.0.17.238 Pier-OmniBrowser version 1.0.17.34 Pier-PersistencyManager version 1.0.16/16 Pier-Squeak-Persistency version 1.0.17.6 Pier-Tests version 1.0.17.104 Refactoring Core version 5 (installed) Refactoring Core version 11 Refactoring Engine version 26 Refactoring Engine version 31 Refactoring Engine version 32 Refactoring Engine version 33 Refactoring Engine version 35 Refactoring Engine version 36 Refactoring Engine version 37 Refactoring Engine version 39 Refactoring Engine version 46 Refactoring Engine version 47 Refactoring Engine version 48 Refactoring Engine version 50 Refactoring Engine version 51 Refactoring Tests version 0 Refactoring Tests version 1 Refactoring Tests version 3 RoelTyper version 0.44 RoelTyper version 0.58 RoelTyper version 0.59 RoelTyper version 0.60 SUnit version 0.33 SUnit-improved version 4.0 SUnit-improved version 4.0.105 SUnit-improved version 4.0.108 SUnit-improved version 4.0.113 SUnit-improved version 4.0.115 SUnit-improved version 4.0.116 SUnit-improved version 4.0.117 (installed) SUnit-improved version 4.0.118 SUnitGUI-improved version 4.0 SUnitGUI-improved version 4.0.48 (installed) SUnitGUI-improved version 4.0.50 Scripting Workspace version 1.22-apha SendTreeExplorer version 1alpha SharedStreams version 1.1 SimpleLog version 1.13 SmaCC version 5 SmaCC version 17 SmaCC version 18 SmaCC runtime version 13 Sport version 2.30 (installed) Sport version 2.31 Squeak dev packages version 0.1 Squeak dev packages version 0.2 Squeak dev packages version 0.3 Squeak dev packages version 0.4 Squeak dev packages version 0.5 Squeak dev packages version 0.6 Squeak dev packages version 0.7 Squeak dev packages version 0.8 Squeak dev packages version 0.9 Squeak dev packages version 0.10 Squeak dev packages version 0.11 Squeak dev packages version 0.12 Squeak dev packages version 0.13 Squeak dev packages version 0.14 Squeak dev packages version 0.15 Squeak dev packages version 0.16 Squeak dev packages version 0.17 Squeak dev packages version 0.18 Squeak dev packages version 0.19 Squeak dev packages version 0.20 Squeak dev packages version 0.21 Squeak dev packages version 0.22 Squeak dev packages version 0.23 Squeak dev packages version 0.24 Squeak dev packages version 0.25 Squeak dev packages version 0.26 Squeak dev packages version 0.27 Squeak dev packages beta version 0.15 Squeak dev packages beta version 0.16 Squeak dev packages beta version 0.17 Squeak dev packages beta version 0.18 Squeak dev packages beta version 0.19 Squeak dev packages beta version 0.20 Squeak dev packages beta version 0.21 Squeak dev packages beta version 0.22 Squeak dev packages beta version 0.24 Squeak dev packages beta version 0.25 Squeak dev packages beta version 0.26 Squeak dev packages beta version 0.27 Squeak dev packages beta version 0.28 Squeak dev packages beta version 0.29 Squeak dev packages beta version 0.30 Squeak dev packages beta version 0.31 Squeak dev packages beta version 0.32 Tamaris version 0.50 Tamaris version 0.51 Tamaris version 0.53 Tamaris version 0.61 TestBrowser version 120 Typeinference version 1 Distributions Pharo packages version 0.1 Pharo packages version 0.2 Pharo packages version 0.3 Pharo packages version 0.4 Pharo packages version 0.5 Pharo packages version 0.6 Pharo packages version 0.7 Pharo packages version 0.8 Pharo packages version 0.9 Fonts FreeType Plus version 0.3 FreeType Plus version 0.4 FreeType Plus version 0.5 FreeType Plus-Base version 0 FreeType Plus-plugins version 0.13 FreeType Plus-plugins version 0.14 FreeType Plus-plugins version 0.15 FreeType Plus-plugins installer version 0.16 FreeType Plus-plugins source version 1 FreeType Plus-plugins source version 2 FreeType Plus-plugins source version 3 Nice fonts version 0.3 Nice fonts version 0.4 Nice fonts version 0.5 Nice fonts version 0.6 X11Fonts version 1-2 Games (installed) BreakOut version 5 Electricity Game version 1.18 (installed) Electricity Game version 1.19 Games version 22-4-2004 (installed) Games version 2005-10-14 GamesTests version 2003-5-2 LifeMorph version 1.0 Quotes version 21 Quotes version 22 SqueakPipes version 20040426 ZigZag version 0.1 Graphics Balloon3D version 0.18 Balloon3D version 0.18.1 Balloon3D version 0.18.2 Balloon3D-Constants version 0.4 Balloon3D-Import version 0.3 Balloon3D-Import version 0.3.1 Balloon3D-Import version 0.3.2 Balloon3D-Kernel version 0.6 Balloon3D-Math version 0.2 Balloon3D-Morphic version 0.1 Balloon3D-Morphic version 0.3 Balloon3D-Morphic version 0.3.1 Balloon3D-Packaging version 0.6 Balloon3D-Plugins version 0.2 Balloon3D-Plugins version 0.2.1 Balloon3D-Plugins version 0.2.2 Balloon3D-Pooh version 0.3 Balloon3D-Pooh version 0.3.1 Balloon3D-Tutorial version 0.4 Balloon3D-Tutorial version 0.4.1 Balloon3D-VRML version 0.2 Balloon3D-VRML version 0.2.1 Balloon3D-VRML version 0.2.2 Balloon3D-Wonderland version 0.4 Balloon3D-WonderlandConstants version 0.2 GraphViz version 41 GraphViz version 41r2 Graphics-External version 0.1 Graphics-Tools version 0.2 Smagick version 0.1 Smagick version 0.2 Smagick version 0.3 Smagick version 0.4 Smagick version 0.5 Group Development Installer version 1.0.181 Installer version 4.1 Installer version 4.2 Installer version 4.3.271 Monticello15 version 1.5+ Monticello16 version 1.6+ Monticello2 version 0.1 Monticello2 version 0.2 Monticello2-Core version 0.182 Monticello2-Core version 0.191 Monticello2-Core version 0.192 (installed) Monticello2-Core version 0.194 Monticello2-OmniBrowser version 0.6 Monticello2-OmniBrowser version 0.8 Monticello2-OmniBrowser version 0.9 (installed) Monticello2-OmniBrowser version 0.17 Monticello2-Squeak version 0.6 Monticello2-Squeak version 0.37 Monticello2-Squeak version 0.55 Monticello2-Squeak version 0.58 (installed) Monticello2-Squeak version 0.62 MonticelloConfigurations version 1.0.42 MonticelloConfigurations version 1.0.43 MonticelloConfigurations version 1.0.44 MonticelloConfigurations-UniversesFix version un.33.fix.1 MonticelloConfigurations-UniversesFix version un.33.fix.2 MonticelloConfigurations-UniversesFix version un.33.fix.3 OB-Monticello version 0.30 Packages-Library version 0 (installed) SARBuilder version 8 Sake version 67 Sake version 78 Sake version 79 (installed) Sake version 82 Sake-Bob version 7 Sake-Packages version 3 (installed) Sake-Packages version 8 Sake-Scheduler version SqCVS version 0.51 SqueakSource version 1.0.6 Universes version 13 Universes version 14 Universes version 15 Universes version 16 Universes version 19 Universes version 23 Universes version 24 Universes version 25 Universes version 28 Universes version 30 Universes version 31 Universes version 32 Universes version 33 Universes version 34 Universes version 35 Universes version 36 Universes version 36.1 Universes version 37 Universes version 38 Universes version 39 Universes version 40 Universes version 41 Universes version 42 Universes version 45 Universes version 47 Universes version 49 (installed) Universes version 50 Universes OmniBrowser version 0.11 Universes OmniBrowser version 0.15 Universes OmniBrowser version 0.16 Universes OmniBrowser version 0.19 Universes OmniBrowser version 0.24 Universes OmniBrowser version 0.26 Universes OmniBrowser version 0.27 Universes OmniBrowser version 0.32 Universes OmniBrowser version 0.34 Universes OmniBrowser version 0.35 Universes OmniBrowser version 0.36 Universes OmniBrowser version 0.37 (installed) Universes OmniBrowser version 0.38 IDE Algernon version 1.0.1 Algernon version 1.1 Algernon version 1.2 KeyBinder version 2004.8.14 KeyBinder version 2004.8.14.18 Keymapping version 4.1 Launcher version 1.0.62 OmniBrowser-Algernon version 0.5 SVI version 0.115 ScriptManager version 0.6 Scripter version 0.5.5 Techo-Shout version 4 Kernel KernelExt-kph version 3.10.4 KernelExt-kph version 3.10.10 KernelExt-kph version 3.10.12 KernelExt-kph version 3.10.20 KernelExt-kph version 3.10.24 KernelExt-kph version 3.10.26 KernelExt-kph version 3.10.29 KernelExt-kph version 3.10.53 KernelExt-kph version 3.10.54+ SystemEditor version 1 SystemEditor version 1+ SystemEditor-Core version 1.155 SystemEditor-Core version 1.156 SystemEditor-Squeak version 1.149 SystemEditor-Squeak version 1.165 Languages and Extensions DynamicBindings version 1.1 DynamicBindings version 1.21 DynamicBindings version 2.7 DynamicProtocols version 0.39 Lisp version 1 Logic expressions version 1.0-alpha Logic expressions version 2.1 Mathematics DHB Numerical Analysis version 1.0 Numerics version 1 Numerics version 2 (installed) Numerics version 4 Units version 1 Units version 3 (installed) Units version 4 Model Extension AST version 0.103 AST version 0.104 AST version 0.113 AST version 136 AST version 137 AST version 139 AST version 140 AST version 141 AST version 142 AST version 144 AST version 145 AST version 146 AST version 147 AST version 149 AST version 150 AST version 151 AST version 153 AST version 154 AST version 155 AST version 157 AST version 158 (installed) AST version 160 AXAnnouncements version 1.0.31 Announcements version 0.5 Announcements version 0.7 ClassSelectorSets version 1 File-Base version 4 File-Base version 6 File-Base version 8 File-Kernel version 4 File-Kernel version 6 File-Kernel version 7 File-Kernel version 8 File-Test version 4 File-Test version 5 FileMan version 0.44 FileMan version 0.85 LambdaMessageSend version 16 Nile version 0.8.0 Nile version 0.8.1 Nile version 0.8.2 Nile version 0.9.0 Nile version 0.9.5 Nile version 1.0.121 Nile version 1.0.122 Nile-Base version 1.0.57 Nile-Base version 1.0.58 Nile-Clients version 1.0.92 Nile-Clients version 1.0.93 Nile-Tests version 1.0.63 Nile-Tests version 1.0.64 Nile-Tests version 1.0.65 Null version 0.7 Null version 0.8 Null version 0.13 (installed) Null version 0.14 Rio version 0.34 Rio version 0.34.14 Rio version 0.34.18 Rio version 0.34.21 Rio version 0.34.22 Rio version 0.34.29 Rio version 0.81.35 Rio version 4 Rio-Kernel version 0.36 Rio-Kernel version 0.63 Rio-Kernel version 0.67 Rio-Kernel version 0.70 Rio-Kernel version 0.71 Rio-Kernel version 0.78 Rio-Kernel version 0.81 SystemChangeNotifier improvements version 1 Toothpick version 0.1 VBRegex version 1.8 (installed) VBRegex version 1.9 Morphs EventInterceptorMorph version 1alpha FractalMorph version 1.2 Magritte-Morph version 1.0.17.42 MemoryUsage version 0.3 MultiColumn List Morph version 6 PieChartMorph version 0.2 Plot Morph version 3 Screen Shot Morph version 0.0.0.1 Screen Shot Morph version 0.0.0.2 WanderingLetters version 22Feb-2.1 WarpSketchMorph version 1.12 Network Celeste version 1.25 Celeste version 2.21 DNS Client version 1.0 DNS Client version 1.1 ExternalWebBrowser version 8 ExternalWebBrowser version 9 Folktale version 0.51 HTML-Parser version 1.5 HTML-Parser version 1.5-2 HTTPClient version 19 IRCBot version 2.1 IRCe version 10.7.6-2 KomHttpServer version 6.2 KomHttpServer version 7.0.4 KomHttpServer version 7.0.30 KomServices version 1.0 KomServices version 1.1.2 KomServices version 1.12 PWS version 0.2gt1 REPLServer version 16 RFB-VNC version 1.0 RemoteFrameBuffer version 34 SoapCore version 1.0b Swazoo version 2.0 Swazoo version 2.1 Swazoo-HTTP version 7 Telnet version 301 Office Diagram Browser version 2004-7-25 MiniToDo version 1.0 MiniToDo version 1.1 OpenOffice Importer version 16 OpenOffice Importer version 17 X11Fonts version 1 Persistence BTree version 28 BTree version 49 GOODS version 71 GOODS version 80 Glorp version 0.4.129 Glorp version 0.4.169 Java Serialization version 1.61 Java Serialization version 1.71 Magma 1.0 client version r40 Magma 1.0 server version r40 Magma 1.0 server version r41 Magma 1.41 version r41.1 Magma Client version 0.9beta Magma Seaside version 0.9beta Magma Seaside-alpha version 0.9.83 Magma Seaside-alpha version 0.9.84 Magma Seaside-alpha version 1.0.86 Magma Server version 0.9beta Magma seasideHelper version 2.8.r40 Magma seasideHelper version 2.8.r40.32 Magma sunit version 1.0.6 Magma sunit version 1.0.8 MinneStore version 2.8 Mysql Driver version 1.13 NetStrings version 1.0 ODBC version 1 ODBC version 2 PostgreSQL Client version 0.9.2 (installed) PostgreSQL Client version 1.0 ROE version 42 SIXX version 0.1h SIXX version 0.3 SIXX version 0.55 SIXX version 0.74 SIXX version 0.82 SIXX version 0.83 (installed) SIXX version 0.95 SPL PDF Library version 1.0-21a SandstoneDb version 113 SqueakDBX version 0.7 SqueakDBX version 0.7.1 (installed) YAXO version 9 Simulation Manzana version 2003-1-21 (installed) SmallDEVS version 2006-12-6 Squeak Distributions Squeak-dev image version 0.96 Squeak-dev image version 0.97 Squeak-dev image version 0.98 Squeak-dev image version 0.105 Squeak-dev image version 0.106 Squeak-dev image version 0.107 Squeak-dev image version 0.108 Squeak-dev image version 0.114 Squeak-dev image version 0.115 Squeak-dev image version 0.116 Squeak-dev image version 0.117 Squeak-dev image version 0.118 Squeak-dev image version 0.120 Squeak-dev image version 0.123 Squeak-dev image version 0.125 Squeak-dev image version 0.126 Squeak-dev image version 0.128 Squeak-dev image version 0.129 Squeak-dev image version 0.130 Squeak-dev image version 0.133 Squeak-dev image version 0.134 Squeak-dev image version 0.137 Squeak-dev image version 0.140 Squeak-dev image version 0.143 Squeak-dev image version 0.144 Squeak-dev image version 0.148 Squeak-dev image version 0.151 Squeak-dev image version 0.154 Squeak-dev image version 0.155 Squeak-dev image version 0.156 Squeak-dev image version 0.158 Squeak-dev image version 0.159 Squeak-dev image version 0.160 Squeak-dev image version 0.161 Squeak-dev image version 0.162 Squeak-dev image version 0.163 Squeak-dev image version 0.164 Squeak-dev image version 0.165 Squeak-dev image version 0.167 System CommandShell version 3.0.5 CommandShell version 4.2.2 CommandShell version 4.2.3 CommandShell version 4.3 Cryptography Team Package version 1.1 Exupery version 0.4 Exupery version 0.10 Exupery version 0.11 Exupery version 0.12 Exupery version 0.12new Exupery version 0.13 Exupery version 0.14 Exupery Development version 0.12 Exupery Development version 0.12r2 Exupery Development version 0.14 Exupery Development version 0.14.1 Exupery Development version 0.14.2 Exupery Development version 0.14.3 FFI version 3.9 FFI version 3.9.1 FFI-Examples version 3.9 FFI-Tests version 3.9 Namespaces version 46 NewCompiler version 117 OSProcess version V4-3-6 OSProcess version V4-3-7 OSProcessIO version 0.1 OSProcessIO version 0.2 OSProcessIO version 0.3 OSProcessPlugin version 4.2.6a OSProcessPlugin version 4.2.7a.1 ProcessSpecific version 1.0.11 ProcessSpecific version 1.0.15 SSH version 1.0 SSL version 1.0 SqueakElib version 1 VMMaker version 3.7b5 VMMaker version 3.8b6 VMMaker version 3.8b6.1 Tools NamespaceTools version 36 PlusTools version 37.3 ToolBuilder-Examples version 4 ToolBuilder-Kernel version 20 ToolBuilder-Kernel version 26 ToolBuilder-Kernel version 27 ToolBuilder-MVC version 12 ToolBuilder-MVC version 14 ToolBuilder-Morphic version 22 ToolBuilder-Morphic version 27 ToolBuilder-Specs version 1 ToolBuilder-Tweak version 23 Uncategorized Cryptography version 0.3 FreeType Plus-plugins source version 0 IRCe version 10.7.6 Mondrian version 0.33 Mondrian version 0.40 Morse Code version 2 Network-HTML version 4 OSProcessPlugin version 4.2.6 ProjectsAsPNG version 1.0 Refactoring Tests version 2 SUnitProgress version 0.2 ScriptManager version 0.4 ShoutWorkspace version 1-tween.1 ShoutWorkspace version 1-tween.2 ShoutWorkspace version 1-tween.3 ShoutWorkspace version 1-tween.4 Speech version 9 Techo-Base version 1 TimeZoneDatabase version 1.2.5 VersionNumber version 1.0 Video and Image Processing version 2.1 User Interface (installed) Diff Tools version 0.51 (installed) Event Enhancements version 0.6 Genie version r67 (installed) Geometry Models version 0.3 Polymorph EventEnhancements version 1.0 Polymorph EventEnhancements version 1.1 Polymorph Geometry version 1.0 Polymorph TaskbarIcons version 1.0 Polymorph ToolBuilder version 1.0 Polymorph ToolBuilder version 1.1 Polymorph Tools Diff version 1.0 Polymorph Tools Diff version 1.1 Polymorph Widgets version 1.0 Polymorph Widgets version 1.1 ToolBuilder integration for UI Themes version 0.51 ToolBuilder integration for UI Themes version 0.52 UI Theme Taskbar Icons version 0.1 UI Themes and Standard Widgets version 0.5 UI Themes and Standard Widgets version 0.53 UI Themes and Standard Widgets version 0.54 UI Themes and Standard Widgets version 0.55 UI Themes and Standard Widgets version 0.56 UI Themes and Standard Widgets version 0.57 UI Themes and Standard Widgets version 0.58 UI Themes and Standard Widgets version 0.59 UI Themes and Standard Widgets version 0.60 UI Themes and Standard Widgets version 0.70 (installed) UI Themes and Standard Widgets version 0.80 Web Development Aida version 5.4 Aida version 5.6 Comet version 2.8.3.28 Magritte-Seaside version 1.0.17.278 Pier version current Pier-Blog version 1.0.17.90 Pier-Commands-RemoveChildren version 1.0.14 Pier-Commands-RemoveChildren version 1.0.14.2 Pier-Design version 1.0.17.5 Pier-Documents version 1.0.17.7 Pier-EditorEnh version 1.0.17.16 Pier-Google version 1.0.17.5 Pier-Jetsam-Environment version 1.0.16.5 Pier-Jetsam-Environment version 1.0.16.7 Pier-LightBox version 1.0.17.6 Pier-Magma version 1.38 Pier-Magma-beta version 1.0.9-alpha.35 Pier-Math version 1.0.17.6 Pier-Media version 1.0.17.1 Pier-Randomizer version 1.0.17.4 Pier-Seaside version 1.0.17.311 Pier-Security version 1.0.17.113 Pier-Setup version 1.0.17.30 Pier-Shout version 1.0.17.8 Pier-Sitemap version 1.0.17.20 Pier-TabbedPagesWidgets version 1.0.14.5 Pier-TagCloud version 1.0.17.11 Pier-Titles version 1.0.17.1 PlotKit version 0.9.1.26 PlotKit version 0.9.1.27 RSRSS2 version 2.8.3.11 Scriptaculous version 2.8.3.242 Seaside version 2.8.3.572 Seaside-PlotKit-beta version 0.9.1.26 Seaside28Jetsam version 2.8.526 Seaside28Jetsam version 2.8.529 Seaside28Jetsam version 2.8.536 Seaside28Jetsam version 2.8.545 Seaside28Jetsam-Blueprint version 0.7.1 Seaside28Jetsam-Supersleight version 0.1 Seaside28Jetsam-Useful-CSS version 0.1 ShoreComponents version 3.10 Squeak web packages version 0.1 Squeak web packages version 0.2 TinyWiki version 1.0.1 Web Development beta Sandcastle version 0 Sandcastle version 1.0 Seaside-Adapters-Core version pmm.2.mcz Seaside-Adapters-Swazoo version pmm.7.mcz-2 Seaside-Adapters-Swazoo version 8 Seaside-Adapters-Swazoo version 9 Seaside-HTML5-beta version 2.8.411.5 Seaside-Squeak-Adapters version pmm.3.mcz Seaside28Jetsam version 2.8.415 Seaside28Jetsam version 2.8.425 SeasideAdaptersCompatibility version pmm.1.mcz ShoreComponents-beta version 3.10
Somehow I managed to get a change sorter window to open too, and here's the list of change sets (there are no changes in "New Changes"):
OBUniverse-Dan Corneanu.38 OB-Regex-lr.9 OB-Enhancements-dr.265 OB-Refactory-lr.113 Monticello2OB-dc.17 OB-Morphic-dr.63 SmallDEVS-Examples-vj.21 SmallDEVS-GUI-ke.51 SmallDEVS-Core-vj.77 SmallDEVS-MyRepository-vj.61 System-Prototypes.7.cs PlotMorph3-dgd.cs Number_class-readFrom.st OB-Standard-dr.377 Refactoring-Core-lr.11 Pinesoft-Tools-Diff-gvc.8 MC2-Squeak-dc.62 SUnitGUI-kph.50 Sake-Packages-kph.8 Universes-dc.50 SIXX-mu.95 XML-Parser-mir.91 Numerics-rdt.29 FixedDecimal-cbc.13 XML-Parser-mir.9 OmniBrowser-dr.432 AST-lr.160 VB-Regex-sd.9 BitArray1 Null-kph.14 PostgresV2-yj.5 NsTileGame-sbw.17 Morphic-Games-asm.1 Games-Sokoban-sbw.9 Sake-Core-kph.82 Pinesoft-Widgets-gvc.324 Array2D-md.1 Pinesoft-EventEnhancements-gvc.3 Monticello2-dc.194 Sport-2.031 NsTileGame-sbw.19 Units-md.4 ConnectorsBasicDistro-nk.12 ConnectorsDemo-nk.5 BroomMorphs-Connectors-nk.2 BroomMorphs-Base-nk.2 ConnectorsGraphLayout-nk.7 ConnectorsTools-nk.21 ConnectorsShapes-nk.29 ConnectorsText-nk.16 Connectors-nk.187 FSM-nk.42 CGPrereqs-nk.8 ConnectorsFor3.8-187.sar Pinesoft-Geometry-gvc.3 SARBuilder.st BreakOut-sd.5 SUnit-kph.117 7179AdvanceTo3dot10dot2 7178MorphicCandidatesForGoCleanup-M7076-klc 7177ReleaseTests-M7076-wiz 7176AdvanceTo3dot10dot2beta New Changes
Here are at least two things you are doing wrong. It looks like you are trying to load many versions of the same package. Only lead ONE version of each package. Almost always, load the most recent one. In my opinion, all the old versions of packages should be deleted from a package universe.
If you are loading Squeak-dev image, you don't need to load anything else. It has all the generally useful and reliable packages that a developer would want. You load just one package, and you get several dozen loaded automatically. That is the magic of PU.
You might find that you need packages not in "Squeak-dev image", in which case you should feel free to load them. Each package in PU is supposed to list all packages that it depends on, so if you load a package and don't get its prerequisite, that is bug in the package definition.
I have found PU to be useful and reliable. On the other hand, I have never loaded more than four or five packages at once, counting "Squeak-dev image" and everything it requires as a single package. When you realize that most of these packages are independently developed and are not always tested together, you will realize that loading 50 packages all at once is a bad idea.
On Sun, Dec 14, 2008 at 6:28 PM, Greg A. Woods; Planix, Inc. woods@planix.ca wrote:
Squeak-dev image version 0.96 Squeak-dev image version 0.97 Squeak-dev image version 0.98 Squeak-dev image version 0.105 Squeak-dev image version 0.106 Squeak-dev image version 0.107 Squeak-dev image version 0.108 Squeak-dev image version 0.114 Squeak-dev image version 0.115 Squeak-dev image version 0.116 Squeak-dev image version 0.117 Squeak-dev image version 0.118 Squeak-dev image version 0.120 Squeak-dev image version 0.123 Squeak-dev image version 0.125 Squeak-dev image version 0.126 Squeak-dev image version 0.128 Squeak-dev image version 0.129 Squeak-dev image version 0.130 Squeak-dev image version 0.133 Squeak-dev image version 0.134 Squeak-dev image version 0.137 Squeak-dev image version 0.140 Squeak-dev image version 0.143 Squeak-dev image version 0.144 Squeak-dev image version 0.148 Squeak-dev image version 0.151 Squeak-dev image version 0.154 Squeak-dev image version 0.155 Squeak-dev image version 0.156 Squeak-dev image version 0.158 Squeak-dev image version 0.159 Squeak-dev image version 0.160 Squeak-dev image version 0.161 Squeak-dev image version 0.162 Squeak-dev image version 0.163 Squeak-dev image version 0.164 Squeak-dev image version 0.165 Squeak-dev image version 0.167
On 14/12/2008, at 21:13, Greg A. Woods; Planix, Inc. wrote:
OK, one more try I thought.....
I took a fresh 3.10.2-7179 image, set my preferences, opened up the Packages Universe, carefully selected the packages I wanted, being extra careful to avoid the ones that I've had problems with already, and started the installs......
Things looked like they had finished fine, and so I tried clicking on the Packages Universe window again to see what it might say and KABOOM! (this all retyped by hand)
***System error handling failed*** Original error: MessageNotUnderstood: LabelMorph class>>contents:. Debugger error: MessageNotUnderstood: LabelMorph class>>contensts:: [] in Debugger class>>openOn:context:label:contents:fullView: {[:ex | self primitiveError: 'Original error: ', title asString, ', Debuge...]} BlockContext>>valueWithPossibleArgs: [] in MethodContext(ContextPart)>>handleSignal: {[(self tempAt: 2) valueWithPossibleArgs: {exception}]} blockContext>>ensure: MethodContext(ContextPart)>>handleSignal: MessageNotUnderstood(Exception)>>signal LabelMorph class(Object)>>doesNotUnderstand: #contents: UIThemeSoftSqueak(UITheme)>>buttonLabelForText: UIThemeSoftSqueak>>buttonLabelForText: UIThemeSoftSqueak(UITheme)>>buttonLabelFor: PluggableButtonMorph>>newLabel PluggableButtonMorph>>label: [] in Debugger>>buttonRowForPreDebugWindow: {[:quad | aButton := PluggableButtonMorph on: aDebugWindow getState:...]} OrderedCollection>>do: Debugger>>buttonRowForPreDebugWindow: Debugger>>buildMorphicNotifierLabelled:message: Debugger>>openNotifierContents:label: [] in Debugger class>>openOn:context:label:contents:fullView: {[debugger := self new process: process controller: controller co...]} BlockContext>>on:do: [] in Debugger class>>openOn:context:label:contents:fullView: {[[debugger := self new process: process controller: controller co...]}
Type CR to enter an emergency evaluator. Type any other character to restart.
Sadly it doesn't look like there's even any obvious hint as to what package might have caused the problem.
My 5 cents of peso (less of 1 euro cent) bet is the cause is into UITheme or the fancy enhanced look. Look for LabelMorph, is not a class of base 3.10, for this class responds to contents: should be subclass of StringMorph (maybe)
Clearly not even a majority of packages in the currently published universe for 3.10.2 have ever actually been tested.
No, and nobody have this duty this days.
Once more the need of a Czar of Packages arise and once more I call for nominate one and a politics for packages.
I see Jerome and you take serious on this , so why not nominate ourselves as "US Marshalls " and fight outlaws :=)
Edgar
On Mon, Dec 15, 2008 at 5:17 AM, Keith Hodges keith_hodges@yahoo.co.uk wrote:
With Sake/Packages, everyone is a czar of packages :-)
Saying "everyone is a czar" is like saying "nobody is a czar".
This is not a packaging tool problem, and it is not a problem that can be solved with packaging tools. Package Universe could in theory be used to solve this problem. (Sake/Packages might be even better - I'm not knocking it - I'm just saying it is not a tool problem.) All that is necessary is to make sure that every combination of things in the PU is tested.
It is relatively easy to create your own Package Universe. The original idea for PU was that each PU would have an owner, and the owner would make sure that everything in the PU worked together. However, what has happened in practice is that there are only a few PUs, one for each version of Squeak, and each PU has a lot of people who put things in it. There is no process for making sure that everything is tested, so some things are tested and some things are not.
-Ralph
Ralph Johnson wrote:
On Mon, Dec 15, 2008 at 5:17 AM, Keith Hodges keith_hodges@yahoo.co.uk wrote:
With Sake/Packages, everyone is a czar of packages :-)
Saying "everyone is a czar" is like saying "nobody is a czar".
No it isnt.
If there are packages and "nobody" is has the power to edit the definitions and the versions that are loaded then it is almost guaranteed not to work, since no one can fix the problems.
"everyone is a czar" means that it can be a genuinely collaborative processes, and everyone who cares is a potential tester and fixer. They are entirely different.
I remember when some thought wiki's would never catch on.
Keith
Keith Hodges escreveu:
Ralph Johnson wrote:
On Mon, Dec 15, 2008 at 5:17 AM, Keith Hodges keith_hodges@yahoo.co.uk wrote:
With Sake/Packages, everyone is a czar of packages :-)
Saying "everyone is a czar" is like saying "nobody is a czar".
No it isnt.
If there are packages and "nobody" is has the power to edit the definitions and the versions that are loaded then it is almost guaranteed not to work, since no one can fix the problems.
"everyone is a czar" means that it can be a genuinely collaborative processes, and everyone who cares is a potential tester and fixer. They are entirely different.
I remember when some thought wiki's would never catch on.
Keith
Here I tend to agree with Ralph. Everyone can cooperate, but cooperation must be coordinated in order to avoid confusion. One model I enjoy (despite its many vices) is the one adopted by fedora.org. A guy must enroll to be maintainer of a package and then, if approved, he is in charge of receiving error notifications and possible cooperations (and I notice that maintainers receive mostly complains and few suggestions/cooperations). If no suggestions shows up, he is in charge of correcting errors. The maintainer may become unresponsive and there are procedures to enroll new maintainers. Either the packages are complaint with fedora.org dispositions (directories, nomenclatures, etc) or they are dropped out (lack of interest in community, obsolescence, etc). The "Fedora Committee" decides what will be in and what will be out of the distribution/image (we can argue about how democratic is the process but I guess we can make the process as democratic as we want in squeak).
Everybody is free to seek for packages out of fedora repositories (freshmeat, sourceforge, etc) but then, if that's the chosen path, the installer is responsible for keep things working.
Another point is that documentation in squeak is really poor. Enrolling maintainers and establishing a standard for documentation would help to make it more popular. Unfortunately browsing methods/messages is not a practical way of doing things if you have to be productive. Also having to find out how things work by the trial and error method is not a solid stand point for convincing your boss that squeak is a fine development tool.
Casimiro de Almeida Barreto wrote:
Keith Hodges escreveu:
Ralph Johnson wrote:
On Mon, Dec 15, 2008 at 5:17 AM, Keith Hodges keith_hodges@yahoo.co.uk wrote:
With Sake/Packages, everyone is a czar of packages :-)
Saying "everyone is a czar" is like saying "nobody is a czar".
No it isnt.
If there are packages and "nobody" is has the power to edit the definitions and the versions that are loaded then it is almost guaranteed not to work, since no one can fix the problems.
"everyone is a czar" means that it can be a genuinely collaborative processes, and everyone who cares is a potential tester and fixer. They are entirely different.
I remember when some thought wiki's would never catch on.
Keith
Here I tend to agree with Ralph. Everyone can cooperate, but cooperation must be coordinated in order to avoid confusion.
At present "one unconfused person" who knows how to get something working is powerless to do so using Universes. Even if you wrote a package, and someone else posted it in the Universe you are powerless to update that entry without ownership of the entry. Having a Czar for universes would be a step forward, at least it would stand a chance, at the moment it is practically locked.
Sake/Packages is open to all to edit the definitions. When we have more than two contributors to Sake/Packages then we can worry about confusion. I dont think there will be much.
I suggested using the packages@lists.squeakfoundation.org list for discussions about Universes and maintaining Sake/Packages etc. I don't think we have had more than 3 posts on the subject there.
One model I enjoy (despite its many vices) is the one adopted by fedora.org. A guy must enroll to be maintainer of a package and then, if approved, he is in charge of receiving error notifications and possible cooperations
So now you have a bottleneck, not just one bottleneck, but Nx bottlenecks. Any attempt to get a combination of 10 packages working can in worst case scenario be halted or hindered by 10 people.
The maintainers of a package do not usually test their packages in all the possible combinations in which it is deployed. The users as a community do.
Sake/Packages enables users to get things working, they can subclass the published definitions and write their own. This then serves as feedback as to what is needed to get something working in a particular context.
To be honest loading stuff is now, at last, really easy, what we need help on now in Sake/Packages now is getting "unloading" stuff working. Many package maintainers don't write #unload methods on their classes.
I regret I myself have never had any success with any of the linux packaging solutions.
regards
Keith
On 15/12/2008, at 12:26, Casimiro de Almeida Barreto wrote:
Here I tend to agree with Ralph. Everyone can cooperate, but cooperation must be coordinated in order to avoid confusion. One model I enjoy (despite its many vices) is the one adopted by fedora.org. A guy must enroll to be maintainer of a package and then, if approved, he is in charge of receiving error notifications and possible cooperations (and I notice that maintainers receive mostly complains and few suggestions/cooperations). If no suggestions shows up, he is in charge of correcting errors. The maintainer may become unresponsive and there are procedures to enroll new maintainers. Either the packages are complaint with fedora.org dispositions (directories, nomenclatures, etc) or they are dropped out (lack of interest in community, obsolescence, etc). The "Fedora Committee" decides what will be in and what will be out of the distribution/image (we can argue about how democratic is the process but I guess we can make the process as democratic as we want in squeak).
This is the same i saying several times.
For democratic procedure, we could add "Czar of Packages" to list of people running for Board in 2009.
If more as one Squeaker run for this duty, who wins the election could be nomitated for one year, same as any running and voted for Board.
Edgar
On 15-Dec-2008, at 7:53 AM, Keith Hodges wrote:
Ralph Johnson wrote:
On Mon, Dec 15, 2008 at 5:17 AM, Keith Hodges <keith_hodges@yahoo.co.uk
wrote:
With Sake/Packages, everyone is a czar of packages :-)
Saying "everyone is a czar" is like saying "nobody is a czar".
No it isnt.
Actually literally it is. Not everyone is capable, let alone inclined to do what you suggest.
Practically there _MUST_ be _ONE_ czar of the _official_ PU for a given release of Squeak (just as there must be one czar for the image content of a given release). Someone must take responsibility if there is to be any real level of quality in the final product. The thing called "Squeak", and the default stuff that ordinary end users can do with it, must be reasonably rock solid and stable and usable.
The Swiki currently contains outright lies about the state of PU for Squeak 3.10. Even most of the basic guidelines of PU are being totally ignored by the current state of the default one for 3.10.2. It is a totally horrendous and completely useless mess so far as I can tell.
At the moment _nobody_ I know will _ever_ even dare try Squeak in this state, not even the programmers, but certainly not anyone more "beginner" than myself.
From a user interface perspective I find SqueakMap infinitely more usable -- if it only contained dependency information and ways to do something with it too.... At least some simple filters can be used to give higher assurance that something might actually work when the user tries it.
Greg A. Woods; Planix, Inc. wrote:
On 15-Dec-2008, at 7:53 AM, Keith Hodges wrote:
Ralph Johnson wrote:
On Mon, Dec 15, 2008 at 5:17 AM, Keith Hodges keith_hodges@yahoo.co.uk wrote:
With Sake/Packages, everyone is a czar of packages :-)
Saying "everyone is a czar" is like saying "nobody is a czar".
No it isnt.
Actually literally it is. Not everyone is capable, let alone inclined to do what you suggest.
Now then lets not get into an argument. I never said that they HAVE to, I said that they CAN.
It only takes ONE person to find that a package doesnt work, and it only takes ONE person to fix it. They don't have to be the same person, and they don't have to be one person.
Anyone who has a problem with any package in Sake/Packages can go onto squeak irc and may find perhaps 2 or 3 people there who will know how to help. They will be able to fix it imedately and once it is fixed then it will available working for everyone.
Squeak has been without an effective packages solution for so long this has become a big deal. Personally I think it is trivially easy to get right, and we dont need a czar at all.
Lets pick an example: Magma:
Magma has 3 installations. Magma client, Magma server, and Magma tester Magma should work in 3.7, 3.8, 3.9, 3.10 , and 3.11(to be), and Pharo (to be), thats 18 different definitions in 6 universes.
So when I fix a bug in magma do I have to contact 6 different czars?
With Sake/Packages I can theoretically manage the package definitions for all 18 in one single image. When a new version of Magma is released it takes less than 1 minute to update the specific definitions for all 8 images. The non-specific 'beta' definitions may simply track "latest" automatically.
Keith
On Dec 15, 2008, at 1:05 PM, Keith Hodges wrote:
So when I fix a bug in magma do I have to contact 6 different czars?
Hopefully not. Hopefully you would not also have to understand the entire packaging system. (Or open your firewall to allow IRC packets to pass...)
- TimJ
Tim Johnson wrote:
On Dec 15, 2008, at 1:05 PM, Keith Hodges wrote:
So when I fix a bug in magma do I have to contact 6 different czars?
Hopefully not. Hopefully you would not also have to understand the entire packaging system. (Or open your firewall to allow IRC packets to pass...)
- TimJ
Oh come on what's with the negativity!
http://www.jwirc.com/chat.html
Keith
On Dec 15, 2008, at 1:33 PM, Keith Hodges wrote:
So when I fix a bug in magma do I have to contact 6 different czars?
Hopefully not. Hopefully you would not also have to understand the entire packaging system. (Or open your firewall to allow IRC packets to pass...)
- TimJ
Oh come on what's with the negativity!
Oh, shame on me, I feel compelled to answer. ;)
IRC is fleeting, temporary, ephemeral... IRC is cliquey... IRC is messy... IRC is addictive... IRC is an old, old friend of mine.
Suppose someone has a question about the packaging system. That person decides to try IRC for a solution. The person asks his question on IRC, and one or more of the following happens:
1) The one person qualified to answer the question is not there. 2) Nobody answers the question. 3) Other conversation ensues, and the person is not able to tear himself away to return to the original problem. 4) The question is answered.
Should the result be #4, that leaves the community with one more person now understanding the answer to the problem. Should another person in the future have the same problem, that person will be stuck trying his luck with the four steps above. Or, that person can try searching Google, and then come up with an IRC log, which he can then try to parse. That is, if the conversation took place in the public channel. If the question was answered via private message, then the exchange which took place on IRC will not help anyone but the person who originally sought the answer.
Please don't think I'm dissing IRC as a whole. I just think it would serve the community best if everyone was on the channel, 24 hours a day, every day, always listening and answering. That's a lot of time spent on IRC. Maybe that's your goal :)
- TimJ
On 15-Dec-2008, at 2:57 PM, Tim Johnson wrote:
Oh, shame on me, I feel compelled to answer. ;)
IRC is fleeting, temporary, ephemeral... IRC is cliquey... IRC is messy... IRC is addictive... IRC is an old, old friend of mine.
Indeed. IRC is not, and never has been, a good place for _anyone_ to just hop in and ask a question, especially not a new user, and probably not even an experienced user. IRC channels are coffee shops, bars, street corners, etc. You go there to converse or proclaim. You don't go there for real information unless you're having a (semi-)private conversation and discussion. IRC channels could be classrooms, but only with external structure and scheduling. People do use IRC channels as a form of Instant Messaging, i.e. in hopes of finding their friend/mentor/support person online and listening, but even with directed IM we all know there are tremendous limitations.
Mailing lists are infinitely better forums for questions, even though they too have their problems.
Should the result be #4, that leaves the community with one more person now understanding the answer to the problem. Should another person in the future have the same problem, that person will be stuck trying his luck with the four steps above.
No, if its a problem with the package definitions, what loads where etc, then when it is fixed it is fixed, for everyone else too.
The point about IRC is about potential speed of turnaround.
Or, that person can try searching Google, and then come up with an IRC log, which he can then try to parse. That is, if the conversation took place in the public channel. If the question was answered via private message, then the exchange which took place on IRC will not help anyone but the person who originally sought the answer.
Please don't think I'm dissing IRC as a whole. I just think it would serve the community best if everyone was on the channel, 24 hours a day, every day, always listening and answering. That's a lot of time spent on IRC.
gad-zukes, you have discovered my secret!
Keith
Maybe that's your goal :)
- TimJ
On 15-Dec-2008, at 2:05 PM, Keith Hodges wrote:
Now then lets not get into an argument. I never said that they HAVE to, I said that they CAN.
:-) Indeed.
However with the current state of affairs it would seem that if you don't build your own PU then you can't have a stable and usable set of packages from which to build your working images with. At my level I'm not even sure how I switch from one PU to another, let alone how I might build my own whole PU! Even calling it a "universe" makes it far too daunting for end users to consider rolling their own. "What, I have to create a whole universe!?!?!?" :-)
Squeak has been without an effective packages solution for so long this has become a big deal.
I would say the main part of the problem is that there is this "new" thing called the Package Universes tool but it really wasn't needed in the first place -- it was a quick blast at an attempt to solve some perceived problems without proper consideration of how those problems could be solved with existing tools and without proper consideration of the effects such a new tool could have on the the thing called "Squeak" and the community that uses it. As such it turns out to be totally un-maintainable and useless.
Unfortunately it is installed as a great big button in the default release and everyone is seemingly told to use it to get stuff they want.
I really think SqueakMap with dependencies would be the right fix. At least with SqueakMap I can filter out stuff that hasn't been "blessed" by its maintainer(s) to work on my current release and that really just leaves the dependencies and conflicts problems. I think end users could pretty much live with SqueakMap if the default filters were set to only show stable packages for the release being run, and if the dependency and conflict handling problems were solved.
Personally if I were in any way responsible for the Squeak 3.10.2 release I'd be removing the Package Universe button from the image and pushing out a new release and update stream _yesterday_.
Lets pick an example: Magma:
Magma has 3 installations. Magma client, Magma server, and Magma tester Magma should work in 3.7, 3.8, 3.9, 3.10 , and 3.11(to be), and Pharo (to be), thats 18 different definitions in 6 universes.
So when I fix a bug in magma do I have to contact 6 different czars?
With the Package Universes way of doing things, IIUC, yes, you really must. Someone has to take responsibility to bless new packages in each release.
Package Universes are the logical equivalent to the pkgsrc/ports systems in the BSD Unix world. Pkgsrc is effectively a set of build and install rules that end users can use to obtain specific versions of packages that have been tested and ported to the OS release they are using. For example in NetBSD pkgsrc the currently "blessed" version of Squeak is still 3.9-final-7067 and it is expected to work on all currently supported releases of NetBSD on any supported hardware platform. If/when someone ports and tests a newer Squeak release to NetBSD _and_ submits an update to the pkgsrc/lang/squeak module then NetBSD users of Squeak will be able to upgrade. Until that time though only the adventuresome who know how to port and test software from scratch will try any newer release of Squeak on NetBSD. This process works for over 7,000 packages that have been ported and tested to work on NetBSD (and a similar amount for FreeBSD "ports"). End users can pick and choose from any or all of those 7,000 packages and have reasonable expectations that they will all install and actually work.
In the BSD world package maintainers who care about their package on a given version of BSD and/or GNU/Linux or Solaris and/or whatever do indeed have to contact each pkgsrc/ports/whatever project to let them know about the new release and perhaps if they really care they'll provide updates to the relevant rules module so that each project can more quickly update their packaging system.
Also, in pkgsrc, for example, maintainers might take responsibility for a given package or sets of packages and watch for updates from the original authors (or even in some cases the pkgsrc maintainer is the author). These maintainers have commit rights to update the relevant pkgsrc rules modules.
See www.pkgsrc.org if you're interested in more detail. There are lots of applicable things that can be learned there -- especially things about dependency and conflict management.
In the Squeak (and Squeak-related smalltalks) world I think SqueakMap could be a better solution for this problem domain in this context, provided of course that the dependency tracking and conflict management problems are solved.
With Sake/Packages I can theoretically manage the package definitions for all 18 in one single image. When a new version of Magma is released it takes less than 1 minute to update the specific definitions for all 8 images. The non-specific 'beta' definitions may simply track "latest" automatically.
Really? I'm not sure how that works. Can you really use one image to test loading into at least 6 separate images? Can you run unit tests from one image in 6 other images?
With SqueakMap, IIUC, you can immediately say which releases you or your beta testers have tested your new packages against. As an unaffiliated user I can choose to turn off the release filter in my SqueakMap interface and see your new version and try it out even if it hasn't been tested on the release I'm using. At least then I know I'm entering new territory on my own.
Greg A. Woods; Planix, Inc. wrote:
On 15-Dec-2008, at 2:05 PM, Keith Hodges wrote:
Now then lets not get into an argument. I never said that they HAVE to, I said that they CAN.
:-) Indeed.
However with the current state of affairs it would seem that if you don't build your own PU then you can't have a stable and usable set of packages from which to build your working images with. At my level I'm not even sure how I switch from one PU to another, let alone how I might build my own whole PU! Even calling it a "universe" makes it far too daunting for end users to consider rolling their own. "What, I have to create a whole universe!?!?!?" :-)
You are preaching to the converted here, I got so frustrated with Package Universes, I re-implemented the whole thing in a weekend, and felt a lot better for it.
The result of that work is Sake/Packages, ok it steals data from Universes, but it is a whole different animal.
Squeak has been without an effective packages solution for so long this has become a big deal.
I would say the main part of the problem is that there is this "new" thing called the Package Universes tool
Its not that new, its been around for a couple of years at least. That's why I figured it had had its day.
but it really wasn't needed in the first place -- it was a quick blast at an attempt to solve some perceived problems without proper consideration of how those problems could be solved with existing tools and without proper consideration of the effects such a new tool could have on the the thing called "Squeak" and the community that uses it. As such it turns out to be totally un-maintainable and useless.
Again, thats why I implemented Sake/Packages.
However what is less clear is how bidirectional integration between SqueakMap and Sake/Packages would be achieved.
Unfortunately it is installed as a great big button in the default release and everyone is seemingly told to use it to get stuff they want.
I am working on a release of 3.10.2 that is specifically designed for building things from. I took that button away.
I really think SqueakMap with dependencies would be the right fix. At least with SqueakMap I can filter out stuff that hasn't been "blessed" by its maintainer(s) to work on my current release and that really just leaves the dependencies and conflicts problems. I think end users could pretty much live with SqueakMap if the default filters were set to only show stable packages for the release being run, and if the dependency and conflict handling problems were solved.
Personally if I were in any way responsible for the Squeak 3.10.2 release I'd be removing the Package Universe button from the image and pushing out a new release and update stream _yesterday_.
or just fixing the universe? Its the 3.10 release folks that are suggesting a czar.
Lets pick an example: Magma:
Magma has 3 installations. Magma client, Magma server, and Magma tester Magma should work in 3.7, 3.8, 3.9, 3.10 , and 3.11(to be), and Pharo (to be), thats 18 different definitions in 6 universes.
So when I fix a bug in magma do I have to contact 6 different czars?
With the Package Universes way of doing things, IIUC, yes, you really must. Someone has to take responsibility to bless new packages in each release.
Package Universes are the logical equivalent to the pkgsrc/ports systems in the BSD Unix world. Pkgsrc is effectively a set of build and install rules that end users can use to obtain specific versions of packages that have been tested and ported to the OS release they are using. For example in NetBSD pkgsrc the currently "blessed" version of Squeak is still 3.9-final-7067 and it is expected to work on all currently supported releases of NetBSD on any supported hardware platform. If/when someone ports and tests a newer Squeak release to NetBSD _and_ submits an update to the pkgsrc/lang/squeak module then NetBSD users of Squeak will be able to upgrade. Until that time though only the adventuresome who know how to port and test software from scratch will try any newer release of Squeak on NetBSD. This process works for over 7,000 packages that have been ported and tested to work on NetBSD (and a similar amount for FreeBSD "ports"). End users can pick and choose from any or all of those 7,000 packages and have reasonable expectations that they will all install and actually work.
I have never got a linux, or bsd installation tool to work for me. I am gifted, I can break anything!
With Sake/Packages I can theoretically manage the package definitions for all 18 in one single image. When a new version of Magma is released it takes less than 1 minute to update the specific definitions for all 8 images. The non-specific 'beta' definitions may simply track "latest" automatically.
Really? I'm not sure how that works. Can you really use one image to test loading into at least 6 separate images?
I said that I can update the definitions for the packages in all images in one place.
Testing is another ball game. The developer of Magma tests it in one image, and I use it in another, thats 2. Soon we will have an auto-building capability that will offer more coverage.
Can you run unit tests from one image in 6 other images?
Magma probably can, since its test suite involves multi-image setups.
With SqueakMap, IIUC, you can immediately say which releases you or your beta testers have tested your new packages against. As an unaffiliated user I can choose to turn off the release filter in my SqueakMap interface and see your new version and try it out even if it hasn't been tested on the release I'm using. At least then I know I'm entering new territory on my own.
Keith
"Greg A. Woods; Planix, Inc." woods@planix.ca writes:
I really think SqueakMap with dependencies would be the right fix. At least with SqueakMap I can filter out stuff that hasn't been "blessed" by its maintainer(s) to work on my current release and that really just leaves the dependencies and conflicts problems. I think end users could pretty much live with SqueakMap if the default filters were set to only show stable packages for the release being run, and if the dependency and conflict handling problems were solved.
Well I have to agree better to have one working things than three different ones. Anyway I can live with SqueakMap without dependencies also because it tells me that it's lacking. As long as I got that information it works quite ok.
Regards Friedrich
On 15/12/2008, at 16:05, Keith Hodges wrote:
So when I fix a bug in magma do I have to contact 6 different czars?
No, only with one, the guy who take the rol with enough consensus of community. Once we have , we know someone care for begginners and not only for people with years of this.
edgar
Edgar J. De Cleene escreveu:
On 15/12/2008, at 16:05, Keith Hodges wrote:
So when I fix a bug in magma do I have to contact 6 different czars?
No, only with one, the guy who take the rol with enough consensus of community. Once we have , we know someone care for begginners and not only for people with years of this.
edgar
It's more than that:
1. We know that someone is supposed to test things for collateral issues. 2. We know that someone will see if conventions established by the community are being followed. 3. We know that someone will be supposed to document the bugs and code evolution. 4. We know that someone will be supposed to care for more subtle problems like dealing with security issues.
I don't like the term czar. In fact not a czar but a person or a group of people in charge of house keeping.
Going to the best known example of large scale free software, Linux didn't catch up until we had:
1. Coherent distributions 2. People in charge of maintaining these distributions 1. Assuring packages are coherent (directories, libraries, terminology, etc) 2. Documenting things 3. Receiving complaints from community (bugzillas and other kinds of complaints) 4. Communicating problems to community 5. Communicating committee decisions to community 6. Communicating community requests to committee
Before that Linux was a toy for students/researchers at universities. "Serious people" used SunOS, later Solaris, AIX, HPUX, Microsoft, MacOS...
Current status is that squeak is still a toy for researchers. If some kind of organization doesn't emerge to deal with less attractive aspects of software life-cycle management, it will hard to find resources to ensure its future.
Current status is that squeak is still a toy for researchers. If some kind of organization doesn't emerge to deal with less attractive aspects of software life-cycle management, it will hard to find resources to ensure its future
We are working on it. Some of us understand these things, but it takes a while to put all of the pieces together. Ask around and you will find people to be generally helpful.
You may find that the answers are already there. The base image you started with is over a year old. Jerome hints that the bug might be caused by a buggy override. If it is then that bug is part of Monticello, not the image. Monticello is a loadable package, and you are able to load a more modern version with the bug fixed.
Monticello1.5 has been available for over a year, and this had "fixing buggy overrides" as one of its first priorities.
Also I think you would be better off building your image with an Installer script, this is more controlled and is what some pro's do See: http://installer.pbwiki.com/Installer
The whole issue with 3.10.2 needing a little extra for successful image building with Installer was the driving force behind the development of MC1.5 and LevelPlayingField.
You can load this via "LevelPlayingField" on http://installer.pbwiki.com/LevelPlayingField How many of these Non-Toy environments you refer to actively support 4 versions back! LevelPlayingField does.
Jerome wrote:
And there were struggles with the method of image maintanence which had shifted at 3.9. Most people will agree that using MC to maintain an image was a problematic choice. My take is that MC is an out of sequence tool. Requiring decisions about packaging to be made too early in the process of writing methods. And patching bugs often cut accross packages. Add that to the fact that MC does not scale well. (Longer and longer times are taken to load larger and larger packages.) And Edgar's time was spent like yours. Learning how to make mistakes and then how to recover from them.
I will write about the 3.11 process in a following email.
Keith
On 15/12/2008, at 18:45, Casimiro de Almeida Barreto wrote:
It's more than that:
We know that someone is supposed to test things for collateral issues. We know that someone will see if conventions established by the community are being followed. We know that someone will be supposed to document the bugs and code evolution. We know that someone will be supposed to care for more subtle problems like dealing with security issues. I don't like the term czar. In fact not a czar but a person or a group of people in charge of house keeping.
Going to the best known example of large scale free software, Linux didn't catch up until we had: Coherent distributions People in charge of maintaining these distributions Assuring packages are coherent (directories, libraries, terminology, etc) Documenting things Receiving complaints from community (bugzillas and other kinds of complaints) Communicating problems to community Communicating committee decisions to community Communicating community requests to committee Before that Linux was a toy for students/researchers at universities. "Serious people" used SunOS, later Solaris, AIX, HPUX, Microsoft, MacOS...
Current status is that squeak is still a toy for researchers. If some kind of organization doesn't emerge to deal with less attractive aspects of software life-cycle management, it will hard to find resources to ensure its future.
Ok. change the name.
Reviewer of Packages ? Pundit ? Marshal ?
Not the same person / group as the Release team. Keith should take a decision. Or he made the release , or he review , but no both. The same person could't be President and Supreme Court Judge !
Edgar
----- Original Message ----- From: "Edgar J. De Cleene" edgardec2001@yahoo.com.ar Date: Tuesday, December 16, 2008 4:27 am Subject: Re: [squeak-dev] not everyone _can_ be a package czar! To: The general-purpose Squeak developers list squeak-dev@lists.squeakfoundation.org
Ok. change the name.
Reviewer of Packages ? Pundit ? Marshal ?
The Bundle Ombudsman. :)
- TimJ
El 12/16/08 2:05 PM, "tjohnson@iwu.edu" tjohnson@iwu.edu escribió:
The Bundle Ombudsman. :)
- TimJ
+1 I need look the exact mean, and like. In Spanish same task could be Defensor del Pueblo :=)
Edgar J. De Cleene wrote:
No, only with one, the guy who take the rol with enough consensus of community. Once we have , we know someone care for begginners and not only for people with years of this.
What exactly stops anyone from being the de-facto packages czar? Given that this is a problem with 3.10 and that you are the de-facto release team leader I see nobody more qualified to address the problem that caused this entire thread than you.
What would the package czar do in this situation? Why can't you do it? If there is any particular reason (or more than one) why you can't do this currently (and this includes both "objective" reasons like access controls as well as "subjective" reasons like community acceptance, lack of quality time etc) I'd like to learn about the issues involved.
Cheers, - Andreas
On 15.12.2008 13:10, Ralph Johnson wrote:
Saying "everyone is a czar" is like saying "nobody is a czar".
This is not a packaging tool problem, and it is not a problem that can be solved with packaging tools. Package Universe could in theory be used to solve this problem. (Sake/Packages might be even better - I'm not knocking it - I'm just saying it is not a tool problem.) All that is necessary is to make sure that every combination of things in the PU is tested.
It is relatively easy to create your own Package Universe. The original idea for PU was that each PU would have an owner, and the owner would make sure that everything in the PU worked together. However, what has happened in practice is that there are only a few PUs, one for each version of Squeak, and each PU has a lot of people who put things in it. There is no process for making sure that everything is tested, so some things are tested and some things are not.
-Ralph
"Testing can show the presence of bugs, but not their absence." (Dijkstra) I dream of the day when code can be read, understood, and audited. Testing will then just confirm that we haven't made any serious blunders. "The more bugs you find during testing, the more bugs there will be in the shipped product. (Trygve) --Trygve
On 15-Dec-2008, at 5:30 AM, Edgar J. De Cleene wrote:
My 5 cents of peso (less of 1 euro cent) bet is the cause is into UITheme or the fancy enhanced look. Look for LabelMorph, is not a class of base 3.10, for this class responds to contents: should be subclass of StringMorph (maybe)
Just clicking through all the change sets in the change sorter shows a LabelMorph class in SmallDEVS-GUI-ke.51.
Here we go again! One more try without SmallDEVS (I think I selected it manually -- i.e. that it was not a dependency of something else I wanted, but I guess I have to carefully look for such dependencies).
squeak-dev@lists.squeakfoundation.org
-
Andreas Raab -
Casimiro de Almeida Barreto -
David T. Lewis -
Edgar J. De Cleene -
Friedrich -
Greg A. Woods; Planix, Inc. -
Keith Hodges -
Ralph Johnson -
Tim Johnson -
tjohnson＠iwu.edu
-
Trygve Reenskaug