Recent

Author Topic: Are We Dead Yet?  (Read 17731 times)

runewalsh

  • Jr. Member
  • **
  • Posts: 70
Are We Dead Yet?
« on: May 08, 2023, 11:49:29 pm »
1) Since the beginning of the year, there have been three times fewer commits into the compiler than in corresponding periods of previous years (350+ vs. 1000+; didn’t count the amount of changes though). At what point should I start to panic? I don’t want to switch to C++...

2) What’s up personally with FPK? He was the only person who occasionally accepted my weird MRs, and looks less active for quite a time.

TRon

  • Hero Member
  • *****
  • Posts: 1841
Re: Are We Dead Yet?
« Reply #1 on: May 09, 2023, 12:23:57 am »
2) What’s up personally with FPK? He was the only person who occasionally accepted my weird MRs, and looks less active for quite a time.
Less active ? what is this then ?  :)

Last time I checked FPK is a human being, and as such deserves some time off once in a while like a holiday, a sabbatical or simply a (short) break from FPC related development. Do note: imho.

jamie

  • Hero Member
  • *****
  • Posts: 5835
Re: Are We Dead Yet?
« Reply #2 on: May 09, 2023, 12:26:21 am »
1) Since the beginning of the year, there have been three times fewer commits into the compiler than in corresponding periods of previous years (350+ vs. 1000+; didn’t count the amount of changes though). At what point should I start to panic? I don’t want to switch to C++...

2) What’s up personally with FPK? He was the only person who occasionally accepted my weird MRs, and looks less active for quite a time.

 This isn't Microsoft Windows you know.

The only true wisdom is knowing you know nothing

TRon

  • Hero Member
  • *****
  • Posts: 1841
Re: Are We Dead Yet?
« Reply #3 on: May 09, 2023, 12:28:29 am »
This isn't Microsoft Windows you know.
Budgetwise that would be nice though  :D

TRon

  • Hero Member
  • *****
  • Posts: 1841
Re: Are We Dead Yet?
« Reply #4 on: May 09, 2023, 12:54:25 am »
@runewalsh:
On a more serious note though you seem to focus on micro optimizations (for a particular cpu and/or particular extensions as well) which is ofc very welcome but I could imagine it does not have the highest priority or that it is rather for someone else to pick up on those.

runewalsh

  • Jr. Member
  • **
  • Posts: 70
Re: Are We Dead Yet?
« Reply #5 on: May 09, 2023, 01:16:30 am »
Less active ? what is this then ?  :)

That’s exactly what I meant, especially if you look at another tab. I’ll wait another half a year or ten years, just got worried and wanted to know what to expect in advance.

you seem to focus on micro optimizations (for a particular cpu and/or particular extensions as well)

It’s not about my obsessions with Index/Compare, but about the observation that the development has slowed down sharply this particular year without any previous trend, maybe there is a reason.

TRon

  • Hero Member
  • *****
  • Posts: 1841
Re: Are We Dead Yet?
« Reply #6 on: May 09, 2023, 01:27:49 am »
It’s not about my obsessions with Index/Compare, but about the observation that the development has slowed down sharply this particular year without any previous trend, maybe there is a reason.
Could be the case that there is a particular reason (I am not aware of it in case it is). I have been following this project practically since its birth and there have been slow times before.

Are you not on the dev ML (fwiw: I'm not) ? E.g. that might be a more appropriate place to discuss your concerns ?

runewalsh

  • Jr. Member
  • **
  • Posts: 70
Re: Are We Dead Yet?
« Reply #7 on: May 09, 2023, 02:18:31 am »
Are you not on the dev ML (fwiw: I'm not) ?

ML as a form of communication has passed me by, so I’ll resort to it only when the timescales mentioned above are exceeded −_−"

dbannon

  • Hero Member
  • *****
  • Posts: 2597
    • tomboy-ng, a rewrite of the classic Tomboy
Re: Are We Dead Yet?
« Reply #8 on: May 09, 2023, 04:34:20 am »
2) What’s up personally with FPK? He was the only person who occasionally accepted my weird MRs,
I have found with Lazarus, may or may not apply to FPC, patches are preferred to pull requests. Don't know why but they are acted upon (one way or another quicker). Git generates an acceptable patch very easily ....

Davo
Lazarus 2, Linux (and reluctantly Win10, OSX)
My Project - https://github.com/tomboy-notes/tomboy-ng and my github - https://github.com/davidbannon

ccrause

  • Hero Member
  • *****
  • Posts: 793
Re: Are We Dead Yet?
« Reply #9 on: May 09, 2023, 11:52:11 am »
Quote
1) Since the beginning of the year, there have been three times fewer commits into the compiler than in corresponding periods of previous years (350+ vs. 1000+; didn’t count the amount of changes though). At what point should I start to panic? I don’t want to switch to C++...

2) What’s up personally with FPK? He was the only person who occasionally accepted my weird MRs, and looks less active for quite a time.
I hate to say this but posts from newbies that look like they are trying to discredit the fpc project and mention an alternative language make me very suspicious ...  :o
where have I seen this before?
Oh I know where I have seen this before..  >:D
I know this will not change your attitude, but for the record: @runewalsh is not a chatterbox on this forum, but an active contributor to the FPC project.

Kays

  • Hero Member
  • *****
  • Posts: 546
  • Whasup!?
    • KaiBurghardt.de
Re: Are We Dead Yet?
« Reply #10 on: May 09, 2023, 02:37:28 pm »
[…] Since the beginning of the year, there have been three times fewer commits into the compiler than in corresponding periods of previous years […]
Maybe on the brink of death. I frequently employ the metric of open issues. There are 1.4k open tickets. About 400 are somehow tagged as features. Considering a team of 20 permanently appointed members, this is a fair number, nothing to worry about.

However, there’s a backlog of roughly 600 issues 5 years or older (for FOSS projects, for commercial products my threshold would be more like 1 year). This is a sizable portion. Frankly, if I hadn’t chosen FPC 15 years ago, I’d refrain from choosing FPC today. By the looks of it, there’s too much a risk FPC will have died in 10+ years.
Yours Sincerely
Kai Burghardt

Eugene Loza

  • Hero Member
  • *****
  • Posts: 622
Re: Are We Dead Yet?
« Reply #11 on: May 09, 2023, 03:39:35 pm »
Quote
However, there’s a backlog of roughly 600 issues 5 years or older
Hmmm... I'm not really sure if this is a representative number.
First, if the issue was not treated 5 years ago, very likely it never will be. In many commercial projects (including those I participate in) such issues are sometimes just closed because "too unimportant" or "of course it's good, but we don't have time for that".
Second, the issue may have been fixed ages ago or no longer makes sense. E.g. in my current hobby project (computer game) I currently have ~300 open issues. I read through all of them around once in a month and close a dozen because I've fixed those and have forgotten I've had another related issue or because I've changed my plans and no longer need this feature. And finally sometimes even though I didn't fix the issue, I cannot reproduce it any longer and close it with comment "hopefully it's gone!". - Something that's not too trivial to do in a solo project, I can only imagine how complex this thing may get in a large team.
« Last Edit: May 09, 2023, 03:42:30 pm by Eugene Loza »
My FOSS games in FreePascal&CastleGameEngine: https://decoherence.itch.io/ (Sources: https://gitlab.com/EugeneLoza)

marcov

  • Administrator
  • Hero Member
  • *
  • Posts: 11076
  • FPC developer.
Re: Are We Dead Yet?
« Reply #12 on: May 09, 2023, 04:06:57 pm »
[…] Since the beginning of the year, there have been three times fewer commits into the compiler than in corresponding periods of previous years […]
Maybe on the brink of death. I frequently employ the metric of open issues. There are 1.4k open tickets. About 400 are somehow tagged as features. Considering a team of 20 permanently appointed members, this is a fair number, nothing to worry about.

However, there’s a backlog of roughly 600 issues 5 years or older (for FOSS projects, for commercial products my threshold would be more like 1 year).

Number of issues is a bad metric, since feature requests that are not immediately requested remain open for somebody interested to do something with them. Quite often they are in areas where the typical interested developer went away or moved on to newer subjects.

Also: both projects have widened in scope considerably through the years.

That is not to say that things are not slowing down. But issue numbers are a tricky metric.
« Last Edit: May 09, 2023, 04:29:57 pm by marcov »

Kays

  • Hero Member
  • *****
  • Posts: 546
  • Whasup!?
    • KaiBurghardt.de
Re: Are We Dead Yet?
« Reply #13 on: May 09, 2023, 04:38:45 pm »
[…] However, there’s a backlog of roughly 600 issues 5 years or older (for FOSS projects, for commercial products my threshold would be more like 1 year).
Number of issues is a bad metric, since feature requests that are not immediately requested remain open for somebody interested to do something with them. […]
It certainly isn’t a gooood metric, I didn’t claim that, but if I were to make an assessment today (“Shall I use FPC?”), I would not necessarily investigate further as to why so many issues are left open. As an outsider I’d draw my own conclusions: “The project lacks focus.” “The project lacks manpower.” “The project is badly programmed.” “The project is poorly managed.” Or, as Runewalsh insinuated, “The project is dead.” :o I’ve been tracking the progress for quite a while now, so I do have some insight; I argue this simply isn’t a feasible option for everyone, though.
Yours Sincerely
Kai Burghardt

marcov

  • Administrator
  • Hero Member
  • *
  • Posts: 11076
  • FPC developer.
Re: Are We Dead Yet?
« Reply #14 on: May 09, 2023, 04:53:05 pm »
but if I were to make an assessment today (“Shall I use FPC?”),

I doubt many will get as far as drawing up statistics on the bugtracker. Usually fitness for purpose is the best metric, and some familiarity (e.g. coming from Delphi)

I usually also look at the example program situation (which admitted, for FPC and Lazarus is not that rosy either, at least as far as pre packaged examples go.

 

TinyPortal © 2005-2018