Just an observation but you sound kind of jaded.
Re: no one is in charge of the dev. team
So you on your own can decide to release the main image of Squeak for everyone else to download? You're not making a lot of sense here. Who decides what goes into the main image? Who decides when it's in beta, and when it's ready for final release? To hear you talk no one is in charge of any of this. Sounds pretty chaotic to me and not very realistic in terms of managing a credible project.
Even though some object to the notion, most of the open source projects I've heard about that are successful have "benevolent dictators" who decide what goes into a release and what doesn't. They take a lot of input, of course, but ultimately one or a few people make the decisions.
Re: code talks
Yeah, I'm beginning to see that. It sounds like the problem is the users who don't want to take the time to dig into the internals don't have a voice at all in your scheme of things. If the developers don't see a use for something in their own projects, no problem. You can just take it out of the main image that everyone downloads, and the users who depend on what you took out can just go take a hike. If this is the prevailing attitude it's just going to turn Squeak into yet another dev. tool that no one besides developers will have an interest in. So much for Smalltalk representing a new computing platform, right? Instead of just serving your own interests, why not take the time to do what it takes to maintain the vision that is there and improve upon it? I'm not speaking just of you when I say this, but of the Squeak developers generally. If it takes a long time to refactor eToys, why not take the time to do it, however long it takes? What's the
rush to get out the next release? That's what I was kind of getting at in my previous post. Why not take the time to do it right? If it means it takes a few years, then it takes a few years.
I understand I could contribute to this, but if none of the other Squeak system developers are interested in the same vision, it's kind of a disincentive to invest the time, because it'll mean I'm carrying the load myself. Plus it may not be appreciated for what it is and get tossed out. This is what I'm alluding to with the "overall vision" thing. I am interested in the overall vision of Squeak as it exists persisting, even though I may not use some of the features. If there's no guiding vision, how is anyone going to have an idea if a contribution to the main image is going to be accepted or not? What's the incentive to make a contribution that's altruistic? In the situation you're talking about, as I see it, it's everyone for themselves. People will just contribute things that they have a personal invested interest in, because at least if it's not accepted, it'll be useful to themselves. It kind of sounds like that's the way things have been go
ing. And it takes me full circle back to what I see as the main problem: the users who are not system developers don't have a voice in the process. Basically they're depending on there *being* a guiding vision, because they don't know enough to have control over the internals. They're just depending on what they're used to continuing to work.
I may very well become a contributor in the future. As I've indicated I'm very interested in the platform. I'm learning about Squeak right now, so I don't think my skills would be that useful in terms of refactoring eToys at this point in time. Maybe in the future.
You mentioned Tweak. I went by a web page for it, and it didn't look like there was a lot to it, at least from the screenshots they were showing. Does it have a complete implementation of eToys or just a subset?
Re: eToys was hacked into Morphic. That's the reason things keep breaking.
In the PC world I've sometimes seen system implementations that are "just enough" to run another technology on top of it. It sounds like this is what happened with Morphic--just enough was implemented to run eToys. I wonder what the expectation was? Was this done by design, and the assumption was that if there were improvements to be made, that the UI would just grow organically out of what was already there? Maybe the original developers didn't have a full implementation of Morphic in mind, but rather just wanted to get the development process started, and hoped that others would just build on top of it. Like I said, I'm just learning about it at this point, so I don't know the internals as well as others on here. Again, I wonder if the reason why stuff keeps breaking, is that the developers who are working with it are trying to impose a model on it that is incompatible with the way it was developed earlier. Maybe whoever did it had other intentions in mind and
the current crop of developers are not seeing it.
I'm finding it difficult to believe that with the design of Smalltalk being so elegant that something like this would be done sloppily. I guess I'm wondering about the history of how this developed.
Re: This is just my opinion
Alright. So I've expressed mine too. :)
---Mark
-------------- Original message --------------
From: Todd Blanchard <tblanchard@mac.com>
The thing you are missing is that no one is in charge of the dev team. The dev team is you, me, and anybody else with time and inclination to build and release software.
Furthermore, there is no "one true vision". We all have visions and are trying to realize them. At the end of the day - code talks. You like eToys? Believe it should live in Squeak forever? Put in the time to clean it up and make it reloadable. Nobody is stopping you. We would all like that. But everybody has his own itch to scratch and finite resources.
The reason people are considering removing eToys is that it made Morphic too complicated and brittle. It slows us down the way it is. It is why things keep getting broken. It wasn't built on top of Morphic - it was hacked into it.
Feel free to put in the time. I think the best approach is to hack it out - clean up Morphic, and then reimplement it on top of a clean Morphic in an architecturally sound way (assuming somebody wants it badly enough to put in the effort).
There is also a completely new UI framework called Tweak that might replace Morphic - and has an implementation of eToys being built on top of it. Which kind of implies that Morphic is an interim solution anyhow and we might as well do the expedient thing.
This is all just my opinion.? I could be wrong.
-Todd Blanchard
I think whoever is in charge of the dev. team needs to be cognizant of the guiding vision for Squeak and insist that the developers of the official distribution stick to it.