While I may have been too enthusiastic about the advent of a Seaside 3.0 version (beta or RC) pretty soon, the Seaside crowd feels the pain of still not shipping:
We need to shorten our cycles and reduce the friction to releasing. We don’t have the manpower to ensure a prefectly bug-free release anyway so all these release levels are a bit pointless. It felt necessary given the degree of change between 2.8 and 3.0 but given that people are using the current alpha in production already, there’s not reason to add 6 months and several more releases into that process.
I agree. The Smalltalk world is seeing part of its resurrection due to the existence and attractiveness of Seaside, and Version 3 Alpha 5 is in productive use for months already, so why not stick a version name on it and ship?
Everybody knows that Software has bugs, and there are products around that are stamped with a version number while being much less stable.
An official Version of 3.0 would signal two things:
- The team considers it stable enough
- Sceptics only need to wait for 3.(0.)1 before they can ship their product which the can start developing now on 3.0
Next week at the VA Smalltalk Forzum Europe 2010 in Stuttgart, I will be talking about some of our experiences in customer projects where they all seem to struggle with the same kinds of problems:
- Many areas of their system’s source code are unknown land where nobody seems to have been before (at least noone from the remaining team)
- Packaging the application is a job nobody seems to understand and everybody has great fear of
- Jobs like configuration management, packaging, deployment and initialization are poorly documented and are always performed along an old, multi-page checklist which includes very “mystic” steps and some code snippets
- It is impossible to load the project code into a fresh image due to poor configuration management
- many more problems having to do with the fact that staffing, funding and motivation of the project were poor over a few years in the early 2000′s due to the “immediate end of life” of the project, which didn’t come and will possibly not come any time soon
I’ll give a short overview of tools and techniques to get back to speed in a “survived” (VA) Smalltalk Project, most of which, ironically, were invented in Smalltalk.
I’ll also show a few utilities that have grown in our customer projects to attack some of the above-mentioned problems.
So if you still haven’t registered, you should really do so now, because this is the unique chance to meet many VA Smalltalk users from all over Europe in one single room and get in touch with Instantiations as well as their partners and customers. Register here for free and join us next tuesday!