A roadmap for 3.9

stéphane ducasse ducasse at iam.unibe.ch
Sun Dec 12 11:03:05 UTC 2004


Hi andreas


you really confuse our point. We are not talking about enabling 
research we are talking of making sure that the system
can evolve which is different.

> Hi Stef -
>
> I have a few comments on your roadmap, some from a general point of 
> view and some from a specific point of view. I'll give you the general 
> points first. When I read your roadmap, it becomes clear that your 
> major goal is to transform Squeak into something that you can do 
> systems research in.

Absolutely wrong.

We thought that without MC in the base image, we cannot manage Sunit 
and other packages. So without MC we are stuck.


> Your priorities show this clearly - from MC in the base image, over 
> RB, traits, new compiler, system dictionary refactoring is everything 
> aimed at defining a playground for systems and software engineering 
> research.

This is not about research this is about been able to have
	security
	mops
	language extensions such as the Tweak primitives without hacking the 
compiler in the corner.
	but apparently everybody prefers patches over simple code

> There is nothing wrong with this particular direction but *please* 
> keep in mind that there are other interests in the Squeak community 
> besides yours and that people with other interests have other 
> priorities. Most importantly for those "other" people is stability of 
> the base system.

	but we know that.

> During the SqC days we have often heard the complaint that Squeak 
> develops to rapidly, but none of that evolution was even remotely as 
> deep and as intrusive as you are proposing now and have been in the 
> last two or three Squeak versions.


> Code that I have been running quite literally for years broke in 3.7, 
> then again in 3.8 for extremely obscure side-effects of some so-called 
> "cleaning".

Give some examples.
Note that during years cruft accumulated in Utilities and in 
SystemDictionary so adding shit on shit and been cumulative is
clearly a sign of long term death. Take 3.4 and browse SystemDictionary 
or Utilities

But again this is quite simple if the community does not:
	- read what we are doing in time (and not one year after) how can we 
react?
	- like what we are doing we will simply work on our own system and be 
certainly much more productive.

> You and collegues need to understand that this represents a major 
> problem for users of Squeak. And while some breakage is acceptable and 
> expected in new versions there should be significant advantages for 
> users in exchange for these breakages (say, like the m17n changes 
> which broke a whole bunch of stuff at the advantage of being able to 
> use international scripts). And I think that being able to do research 
> in software engineering is generally NOT considered a significant 
> advantage for those users.

What is fun is that the motto of alan is inventing the future, but how 
can you invent something by only patching
little stuff there and there. Ask nathanael how time it would have 
taken him to implement traits in a clean Squeak
vs. the one he had. He told me that he spent all his time fixing the 
system, the compiler.... and I trust him since
he is excellent.

This is by not changing (even if we are aware that this can have impact 
and that we want to limit the impact of other others)
that Squeak will improve. Look at new mop, new ide, new ideas are 
coming from java or other languages and this is not
because we have some cool assets such as seaside or croquet that this 
will change radically the problem. If the system is bloated then it 
gets into your way to create new assets.

> It is important that you understand this issue. When I looked at 3.8 
> for porting Tweak to it I was seriously appalled by many of changes 
> that have crept into the system. Proliferation of base classes has 
> become a significant problem over the last versions and it won't get 
> any better with more extensions to the base image. Something that -in 
> lack of a better word- I will call "random reclassification" of 
> methods (quite honestly I do not see any difference in putting methods 
> into SystemDictionary vs. SmalltalkImage except that

What I like with you is that the way you reduce that. I spent a lot of 
time trying to remove junk and classify a bit the mess in 
systemDictionary. SystemDictionary is a namespace and not a junk place 
but I got the message, all the time I spent cleaning that was not  for 
fun or for research point of view just because each time I opened the 
system I nearly fainted and because I was fed up  to say to my students 
"do not look at that this is crap" because else they would not 
understand that we would like smalltalk.

> a) the name SmalltalkImage sucks

Good to hear after all the discussions that we got some months ago.

> and b) you now have to guess which place to look at) does not help to 
> improve the situation.

Because we had to do it slowly to avoid to break.
Now if this wonderful community would have CLEARLY state ok let us 
clean it for real then this would be fixed.
But the fact that we have to do it small pieces by small pieces make 
everything harder for us too.

> *Please* understand that everytime you move a method which has been in 
> a particular place for a number of years you are potentially breaking 
> dozens of packages. *Please* try to understand that this is a major 
> problem for anyone using such methods.

But you have deprecated methods so normally the transition should be 
smoother.

> In this light ...

>> - MC in the base image
>
> is a grave mistake, made worse by the lack of any consistent 
> argumentation why exactly MC would be needed in the base image 
> (compared to a package loaded into full).

Because we cannot manage the in image packages. So in that case we 
should not have in image packages. But nobody
manage the image anymore so this is even simpler.

>> In the past we created SmalltalkImage out of SystemDictionary in the 
>> hope to have SystemDictionary be a nice and simple namespace class.
>> Lot of stuff has been put in coherent places (SystemNavigation, 
>> SpaceTally,
>> Changeset....) But this will not work since people get used to add
>> stuff there and Smalltalk is a cool name. So the new proposal is the
>> following one and is partly implemented. Note however that it will
>> require from us a lot of work so if you are against it please say it
>> and we will only do it for us.
>
> I wouldn't say I'm against it but I'd say that I reserve the right to 
> wait until you've got something to look at before making any 
> judgement. I won't give you card blanche since I am not necessarily 
> convinced that I will like what I see.

But now we will have a discussion in the group here to see what we will 
be doing.
And I'm not convinced that we will continue to help Squeak. especially 
if this is to get this kind of message after
all the energy that we spent to pay attention to other people.

Because andreas you have also to understand that this is much more bold 
and rewarding to build hyper new fancy
stuff than but changing the color of the facade (I'm exaggerating here 
of course) is much easier that renovating the fundations.

>> We plan to proceed that way:
> [... snipped ...]
>
> And I look forward to seeing what you're doing but at the same time I 
> don't see any compelling reasons why we would have to decide today 
> whether these changes would need to be in (base) Squeak 3.9.

But we do not ask for a card blanche, we ask for an analysis and a 
consensus, because we will not
spent time on something that one guy like you can say after 6 months of 
work: oh this is just a reclassification of methods
Come on are you serious?
Is it the way that this community wants to work?


Stef

>
> Cheers,
>  - Andreas
>
> ----- Original Message ----- From: "stéphane ducasse" 
> <ducasse at iam.unibe.ch>
> To: "The general-purpose Squeak developers list" 
> <squeak-dev at lists.squeakfoundation.org>
> Sent: Thursday, December 09, 2004 4:38 AM
> Subject: A roadmap for 3.9
>
>
>> hi all
>>
>> I have been discussing with marcus/alex and also thinking about that 
>> after my remarks on andreas remarks related to
>> changes, interfaces/ cleaning vs changes. So to avoid get frustration 
>> from all the parts and to state clearly what is our
>> plan for 3.9 we would like you to read the following and let us know. 
>> Depending on the reactions, we (the guys from berne) may decide to 
>> produce our own stream, because we must work too ;). But we 
>> understand that other people cannot deal with a system changing. But 
>> in that case we should draw the consequences.
>>
>> Our part that we will work on for 3.9 is to get the system improved 
>> from a developers point of view. We
>> think that having a good foundation is important for everybody.
>>
>> The overall idea is to slowly improve one aspect of Squeak: The idea 
>> to have a System to build the next
>> System with.
>>
>> This has lots of consquences, one is that you want the system to be 
>> in a shape that e.g., students that are not yet
>> complete Squeak experts can start to do experiments with it on a 
>> quite deep level, e.g., change the compiler or
>> add a feature to the Browser.  Another aspect is to have a System for 
>> research. The Traits project has been hugely
>> successful, but it has shown that Squeak does have real problems if 
>> you use it for stuff like that. If Squeak is supposed
>> to be the system of choice for research like this, we need to fix 
>> those problems, we need to make sure that the
>> lessons learned from those projects get fed back into the system. 
>> (One example is the change-notification that was added to 3.7, it is 
>> a direct result of the Traits project)
>>
>> All this does not mean that 3.9 will only be about that stuff: We 
>> surely want to see e.g. the work of Diego beeing
>> included, also any changes that other people would need are welcome.
>>
>> For our perspective here what we would like to have: we put in (p) 
>> the code that  will be in external packages,
>> which likely will be "full".
>>
>> - services
>> - new preferences pane
>> - OB (p?) + browseUnit new version
>> - MC in the base image
>> - rbengine (p)
>> - shout (p)
>> - SqueakMap II (yes we want it)
>> - eCompletion or another package
>> - keybinding
>> - traits
>> - new compiler framework of anthony adapted by marcus to produce
>> not full block. Note that for Etoy the old compiler will still be in 
>> the image so from that point of view
>> it should not have an impact.
>> - refactoring of systemDictionary. (see below this point)
>>
>> We are sure that we forgot something but this is what we have on top 
>> of my head.
>>
>> Now for systemDictionary here is the situation and here is where we 
>> would like to be after 3.9
>>
>> In the past we created SmalltalkImage out of SystemDictionary in the 
>> hope to have SystemDictionary be a nice
>> and simple namespace class. Lot of stuff has been put in coherent 
>> places (SystemNavigation, SpaceTally, Changeset....)
>> But this will not work since people get used to add stuff there and 
>> Smalltalk is a cool name.
>> So the new proposal is the following one and is partly implemented. 
>> Note however that it will require
>> from us a lot of work so if you are against it please say it and we 
>> will only do it for us.
>>
>> from a user point of view we want to have:
>> only one class: SmalltalkImage/SystDict for all the image related 
>> services
>> (VM pluggins, ....) but the namespace behavior will be delegated to 
>> Namespace.
>>
>>        All the previous code referencing Smalltalk will work but with 
>> deprecation for the namespace interface.
>>
>> From the language programmer point of view:
>> there will be a namespace that can be used for experimentation
>> this class will replace the environment class of dan that is now 
>> obsolete
>>
>> We plan to proceed that way:
>> - create a class namespace with a new interface (done)
>> - delegate from systemDictionary to this new class (done but not 
>> published)
>> - deprecate all the namespace method of systemDictionary (but we 
>> likely want to keep them for some time)
>> - fix all the in image senders of Smalltalk as a namespace to use the 
>> new class (using self environment for example).
>> (partly done over the previous years)
>> - merge back SmalltalkImage into SystemDictionary: the idea here is 
>> to not have SystemDict as a subclass of Dictionary
>> - have Smalltalk pointing to an instance of this new entity, so that 
>> everybody is happy.
>>
>>
>>
>> Stef and Marcus
>>
>>
>>
>
>




More information about the Squeak-dev mailing list