[Cuis-dev] Creating a stable distribution to support application development

Gerald Klix cuis.01 at klix.ch
Sat Aug 7 09:12:54 PDT 2021


Hi David,

You could add tags or create branches in
your own checkout of Cuis.

The same could be done by Juan for the Cuis repo.
I am sure if we ask Juan to tag stable Cuis revisions,
he will tell us that Cuis should be stable and consistent in
each an every revision. In my experience it is stable
in every revision, but the documentation is sometimes
not up to date.


Shameless advertising:

I am still trying to provide stable releases with Haver 
(http://haver.klix.ch), which also provides a stable
Cuis image including a VM with the VectorGrahics plugin.
You are all invited to join the effort.
Please drop me note if you want to test Haver,
I neglected updating Haver releases, mostly
because I was not able to create 32-bit
Raspberry-OS VMs. This issue has been resolved.

David, you might notice that I ported one of your projects
to Haver's module system
(https://hg.sr.ht/~cy-de-fect/HaverOnCuis/browse/haver/worlds).
The Haver-distribution also includes your LR1 parser
as a snapshot.


Best Regards,

Gerald



On 8/7/21 5:47 PM, Joseph Turco via Cuis-dev wrote:
> Good points. This is actually an issue with Pharo as well. On the topic of books, The "squeak by example book" is not updated for the latest version, and when following the quinto game, things don't line up sometimes, so this seems to be something different smalltalk images suffer from. I came to cuis for its simplicity, and I can't really appreciate the idea of a "stable" version as I'm new, but Pharo threw me off with how fast it changes (also super confusing interface but that's off-topic).
> 
> Regards,
> 
> Joseph Turco
> 
> Aug 7, 2021 8:37:16 AM Nicola Mingotti via Cuis-dev <cuis-dev at lists.cuis.st>:
> 
>>
>> Hi David,
>>
>> As applications developer who is trying to use Cuis, to me this sounds as an important subject.
>>
>> You propose some kind of Cuis LTS edition, or Cuis Stable.
>>
>> I think this would make also easier to write documentation, in the sense that, for
>> example, the CuisBook could be a referred to the LTS edition. This would ensure that whatever
>> is written in the book has been tested and it is not going to fail.
>>
>> I have no idea how much effort it would take to maintain it.
>>
>>
>> bye
>> Nicola
>>
>>
>>
>> On 8/7/21 1:25 PM, David Faitelson via Cuis-dev wrote:
>> Dear all,
>>
>> First I would like to say a big big thank you for the amazing work you put into Cuis. I think it is one of the best development platforms I had the pleasure of working with (it's closest competitor being only the original Smalltalk-80 environment 🙂).
>>
>> To quote C.A.R Hoare, "Here is a language so far ahead of its time, that it was not only an improvement on its predecessors, but also on nearly all its successors."
>>
>> In the last few years I have developed several non trivial applications on top of Cuis and Pharo: A parser generator and IDE, an educational environment for studying Turing machines, a compiler and IDE for a parallel programming language for Arduino boards, an educational math game, and others.
>>
>> So my interest in Cuis is more as a development platform. And I think that ultimately this is what we all want. That more and more people will use Cuis to develop their systems.
>>
>> However, at the moment (correct me if I'm wrong), there is no stable distribution that one can base their application development on. I can clone the latest repository and packages, and freeze a snapshot, but once development moves on I cannot simply update my system to the latest version, because bug fixes, new features, and incompatible changes are mixed in the same development stream.
>>
>> What I think would be very helpful is to have a stable distribution of the core and associated packages that would only change to fix bugs without affecting the interfaces.
>>
>> I am more than willing to contribute my time to develop this concept, and I have a few ideas about how to do this, but I hope that I can find other people that think this is important and are interested in cooperating on this.
>>
>> What do you think? is this worthwhile? what would be a good way to proceed?
>>
>> Cheers,
>> David
>>
>> *Dr. **David Faitelson*
>>
>> *Senior Lecturer*
>>
>> *|* Office: +972-3-7688782 *|* Mobile: +972-52-6568896
>>
>> Mivtza Kadesh St., Tel-Aviv 38
>>
>> www.afeka.ac.il[http://www.afeka.ac.il/]
>>
>>   [cid:part2.FFE6D3D4.91FCC1EC at gmail.com]
>>
>>
>>
>>
>> --
>> Cuis-dev mailing list
>> Cuis-dev at lists.cuis.st
>> https://lists.cuis.st/mailman/listinfo/cuis-dev
> 
> 


More information about the Cuis-dev mailing list