I think we���re in a kind of in-between path:
- the bootstrap work but still at its early days
- the main development still copes with Morphic. Spec2 is a high objective and is sponsored��� So lots of code duplication, as the tools and frameworks.
But also progress and again so few real manpower��� It seems to be more reasonable to imagine iterative bootstrap of new images. But this is just a feeling and right now people are so busy that I think we have to be patient for that.I don���t know but that might be kind of an obliged path so as to reduce drastically later. I hope so too.
Also, Hilaire, I can see you struggle with DrGEO to keep up to date. I���ve no such project, so I can just imagine your frustration, especially since DrGEO UI had to rely on Morphic from long time (To me, this must be the hardest projects to maintain here).
For now, I can bear some hundreds MB as anyway all open softwares on my OS (just crazy by the way). Just out of curiosity, I opened an Activity Monitor (see pic 1).
Wow, I was amazed that pharo occupies 2,51Go (the processor is ok, around 3% in idle). But WOW, I work in this image for about two weeks never turning off my computer, always suspend with Pharo and everything opened (yes I���m this kind of guy ^^) and this is actually extraordinary stable.
I���ve as many workspaces as I did visualization experiments on COVID��� So thousands of record every days, and visualizations etc��� Compared to other softwares on the list, pharo does not have to blush.
I save the image, it is 170MB on disk. 20MB more than my last version (with is normal since I keep more data in it). I reopen.
All is still available (at least the data in playgrounds), and the occupied memory is 228MB (pic 2). Ok, I could have done a garbage collector���
But still I���m impressed on the stability, really ! So I���d like to sincerely congratulates the dev for that. Please continue your hard work !
I also wish that the infrastructure will soon allow to build real and better images, within a contained complexity. Just some questions maybe :
- what would be a realistic estimation of having �� minimal images from bootstrap�� usable in production ? 2y ?
- What about the UI plans here, Full removal of Morphic ? SPEC2 then specific backends (Bloc / Native) ? Is it possible/planned to have a programmable UI, like a REST Interface that would be compatible with Spec2 ?
C��drick
After reopening (like one of my safari tab !).
As a matter of fact it is interesting to compare the number of classes in the last releases of the descendants of Smalltalk 80:
* Cuis5 (480 classes),
* Squeak 5.2 (2713 classes)
* Pharo8 (*9084 *classes)
I don't think the last Pharo vision statement:
/A system with robust abstractions that a single person can understand./
is in anyway realistic.
Now, will Pharo9 reach the sky with *10'000* classes? Very likely if you observe the graph below:
Hilaire
Le 13/04/2020 �� 14:30, C��drick B��ler a ��crit :
Just a reflexion that I���d like to share here.
When Hilaire talks about complexity. I have to agree too, even if I like the progress under the hood, and explored spaces like GToolkit.
--
Dr. Geo
http://drgeo.eu