# Release strategy for juliac?
**URL:** https://discourse.julialang.org/t/release-strategy-for-juliac/112563
**Category:** Community
**Tags:** binarybuilder, juliac
**Created:** [April 5, 2024, 9:24am UTC](https://discourse.julialang.org/t/release-strategy-for-juliac/112563 "2024-04-05T09:24:03Z")
**Posts on this page:** 11
**Page:** 2
### Author:  [@sylvaticus](https://discourse.julialang.org/u/sylvaticus)
#### Post date: [April 5, 2024, 8:03pm UTC](https://discourse.julialang.org/t/release-strategy-for-juliac/112563/21 "2024-04-05T20:03:04Z")
Sorry for beeing very newbby, but what is `juliac`?
Googling I saw that it was a CLI util part of PackageCompiler, but it got removed long time ago…
---
### Author:  [@mbauman](https://discourse.julialang.org/u/mbauman)
#### Post date: [April 5, 2024, 8:06pm UTC](https://discourse.julialang.org/t/release-strategy-for-juliac/112563/22 "2024-04-05T20:06:57Z")
It doesn’t exist — not yet. It’s a mythical Julia frontend that spits out standalone executables. And also spits out flying unicorns. And brings about world peace.
---
### Author:  [@oheil](https://discourse.julialang.org/u/oheil)
#### Post date: [April 5, 2024, 8:40pm UTC](https://discourse.julialang.org/t/release-strategy-for-juliac/112563/23 "2024-04-05T20:40:33Z")
I am not trying to put pressure on the development. I try to convince people that this is a major step in the evolution of Julia which should be considered to be published accordingly. For this to be successful it needs to be well tested.
> [@mbauman](#):
>
> all while also not being involved in its development
Well, it’s beyond my capabilities, but I try to help Julia in general, e.g. by working on some packages and by moderating Forem, as my time allows it.
> [@mbauman](#):
>
> name itself could be held back a bit?
What do you mean? It wasn’t me talking about it.
> [@mbauman](#):
>
> but you’re are also excited to beta-test it
Yes, because this would be within my capabilities, so I would do some work on beta test it. Which is work too, not only fun, if done the right way.
Your answer sheds a strange light on my motivation, or do I read it wrong? Again, it’s not about getting faster to `juliac` for myself, it’s about bringing Julia, the _new_ Julia, to the people who yet don’t want to or can’t use it. The current niche only accounts for 1% of the developers out there. There are 99% others to reach for. Roughly 🙂
---
### Author:  [@\_bernhard](https://discourse.julialang.org/u/_bernhard)
#### Post date: [April 5, 2024, 8:47pm UTC](https://discourse.julialang.org/t/release-strategy-for-juliac/112563/24 "2024-04-05T20:47:42Z")
The point is, at the beginning of beta-testing, by definition, it won’t be well-tested yet - that’s what the beta-testing is for.
However, to beta-test it, it needs to be released, which is, at least superficially, a conundrum… ^^
---
### Author:  [@mbauman](https://discourse.julialang.org/u/mbauman)
#### Post date: [April 5, 2024, 8:51pm UTC](https://discourse.julialang.org/t/release-strategy-for-juliac/112563/25 "2024-04-05T20:51:41Z")
You’re reading it wrong. I’m not casting aspersions at your motivations at all, I just don’t see how all your desires can _possibly_ coexist. You’re so excited that at a mention of it you start a whole thread about a release strategy! I mean, I’m excited, too, but this topic itself is a splash! And that’s the point — folks (and not just you) **are** hungry for this. So the moment it (or anything like it) is available, folks are going to be excitedly rabbling on about it… and I’m not sure what to do about that. My question was in earnest — maybe that’s one possible solution?
Yes, I regret name-dropping `juliac` in the first place… but anyone can look at the PRs (like Pall did) on Julia’s repo and follow along with the work, false starts and all, even before its ready.
---
### Author:  [@oheil](https://discourse.julialang.org/u/oheil)
#### Post date: [April 5, 2024, 8:55pm UTC](https://discourse.julialang.org/t/release-strategy-for-juliac/112563/26 "2024-04-05T20:55:42Z")
> [@\_bernhard](#):
>
> to beta-test it, it needs to be released
To beta-test it, it needs to be made available to beta testers. After beta tests are finished and everything is fine, a software is released (to the public). Or is anybody thinking, a Julia release is not tested and with release beta test ist starting? I don’t know.
I would like to discuss, if a `juliac` needs special attention before and during release? Is it as important as I think? Or is it just another release of Julia like before after 1.0?
---
### Author:  [@oheil](https://discourse.julialang.org/u/oheil)
#### Post date: [April 5, 2024, 8:59pm UTC](https://discourse.julialang.org/t/release-strategy-for-juliac/112563/27 "2024-04-05T20:59:01Z")
> [@mbauman](#):
>
> You’re reading it wrong
Fine than.
> [@mbauman](#):
>
> So the moment it (or anything like it) is available, folks are going to be excitedly rabbling on about it… and I’m not sure what to do about that.
That’s the point. It should be concerted in a good marketing way, in my opinion of course. I am neither sure so I started this thread asking people what they think.
---
### Author:  [@\_bernhard](https://discourse.julialang.org/u/_bernhard)
#### Post date: [April 5, 2024, 9:04pm UTC](https://discourse.julialang.org/t/release-strategy-for-juliac/112563/28 "2024-04-05T21:04:22Z")
However, Julia versions in beta or even alpha stage are still accessible to the public, so are nightly builds. Beta testers in julia are actually “the public”.
But if all you’re saying is, that a stable version should only be tagged after enough testing, then I agree.
This, however, is not aligning with your idea of a concerted “release with a bang” anymore. It’s actually quite the opposite…
---
### Author:  [@jakobnissen](https://discourse.julialang.org/u/jakobnissen)
#### Post date: [April 7, 2024, 8:37am UTC](https://discourse.julialang.org/t/release-strategy-for-juliac/112563/29 "2024-04-07T08:37:44Z")
Honestly I’m just excited that juliac is being worked on. Even without being involved in the work behind it, I can see that it must take some very hard design and engineering work to realize - and I’m curious how practical juliac will end up being.
For example: What do you do about type instability? Error if juliac detects any? If so, how often is Base itself type unstable in a way users can’t easily work around? Does the compiler giving up due to recursion cause annoying issues? How often will users find that an inferrable program stops inferring due to minor, seemingly insignificant changes that cause some compiler threshold to be reached? What’s the compilation speed going to be like? Will there be such a thing as a Julia DLL, and how will that work with generic code? Etc etc etc.
These are legit hard problems to solve, and the way they’re solved does matter to how usable it’ll end up being. It’s going to be exciting to see all that play out, and at the current stage I don’t have any clue how ergonomic a statically compiled Julia will feel. In the light of that, talking about a release strategy is too early.
---
### Author:  [@vettert](https://discourse.julialang.org/u/vettert)
#### Post date: [April 7, 2024, 10:53am UTC](https://discourse.julialang.org/t/release-strategy-for-juliac/112563/30 "2024-04-07T10:53:41Z")
(A link to something explaining what juliac is would be appreciated by me and maybe others)
---
### Author:  [@oheil](https://discourse.julialang.org/u/oheil)
#### Post date: [April 7, 2024, 10:56am UTC](https://discourse.julialang.org/t/release-strategy-for-juliac/112563/31 "2024-04-07T10:56:46Z")
> [@vettert](#):
>
> (A link to something explaining what juliac is would be appreciated by me and maybe others)
See @Palli 's post above: [Release strategy for juliac? - #9 by Palli](https://discourse.julialang.org/t/release-strategy-for-juliac/112563/9)
[Previous page](https://discourse.julialang.org/t/release-strategy-for-juliac/112563.md?page=1)