Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Perl 6, 6.d, Diwali, Raku, Rakudo, Moar, NQP, Star, * #2627

Open
AlexDaniel opened this Issue Jan 19, 2019 · 11 comments

Comments

Projects
None yet
10 participants
@AlexDaniel
Copy link
Member

AlexDaniel commented Jan 19, 2019

This ticket is probably more suitable for https://github.com/perl6/user-experience/ but I want to give it a spin here for a moment.

See https://twitter.com/rjha94/status/1086572397812019203 for a very good point.

I think there are ways we can simplify things.

  1. Diwali – What exactly are we achieving by giving a name to every language release? I can understand the historical reasoning behind “Christmas”, but why should we continue with names? There was a long discussion on what “D…” name to pick, can't we just skip this unnecessary step? I propose the next release be just 6.e (that's it).

  2. Rakudo * – I have seen it being written like this, please no. Let it be “Rakudo Star” without alternative ways to write it.

  3. Rakudo Star – the importance of having Rakudo Star is diminishing because most linux distros already come with a relatively recent rakudo package (see https://repology.org/metapackage/perl6/versions and perl6/user-experience#29). So what exactly is the problem that we are trying to solve with Rakudo Star? Is it packaging for Windows? If so, what if we call it “Rakudo for Windows”?

  4. NQP – what are the chances that some other compiler will use NQP? What if we call it just “Rakudo Middleware”, or something along the lines? So that it's clear that it is part of rakudo, just separated a bit.

I may be very wrong on some of the points, so please feel free to comment.

@AlexDaniel

This comment has been minimized.

Copy link
Member Author

AlexDaniel commented Jan 19, 2019

By the way, each Rakudo release used to have a name too!

@jonathanstowe

This comment has been minimized.

Copy link
Contributor

jonathanstowe commented Jan 19, 2019

I'm with you on 1,2 and 3 but I quite like the idea of nqp having a life of its own.

@Altai-man

This comment has been minimized.

Copy link
Contributor

Altai-man commented Jan 21, 2019

Count me +1 for 1, 2 and 3.

@patzim

This comment has been minimized.

Copy link
Contributor

patzim commented Jan 21, 2019

I especially agree on 3. Some more elaboration:
I have for a long time doubted the need for a Rakudo Star. People should be taught and encouraged to find their way around the module ecosystem. Rakudo Star is doing the opposite.

On the other hand having a package that includes a compiled rakudo and zef is helpful, not only on Windows. Such a slim package could simply be marketed as our binary release.
Node.js, Python and Ruby all do it like that. Both their source and binary releases include npm, pip and gem respectively.

@lucasbuchala

This comment has been minimized.

Copy link
Collaborator

lucasbuchala commented Jan 21, 2019

I think there are ways we can simplify things.

Adding another option:

  1. Revoke/retract the "Raku" alias.

I'm not saying this is a good idea or not. I just want to survey the feeling about this in thumbs-up/down.

Link: perl6/user-experience#31

@cygx

This comment has been minimized.

Copy link
Contributor

cygx commented Jan 21, 2019

  1. I don't see anything wrong with keeping version names, though they probably shouldn't feature too prominently (more of an easteregg instead of the Official Name - as a user, what I care about is use v6.e)

  2. 'Rakudo Star' in full and 'R*' in casual conversation

  3. As an anecdote, I've never used R*. Some easy way to get some batteries delivered might be nice, though (meta-packages?)

  4. In the old days, the equivalent was known as 'Parrot Compiler Tools', so perhaps 'Rakudo Compiler Tools' could be another possibility

  5. I never was a fan of the renaming/aliasing, but now that it's happened, no idea how best to proceed. Fact is, now that Zoffix has moved on, unless someone else picks up the torch, I suspect the alias will die a quiet death on its own...

@AlexDaniel

This comment has been minimized.

Copy link
Member Author

AlexDaniel commented Jan 21, 2019

Revoke/retract the "Raku" alias.

Raku/Perl 6 debate is intentionally left out of this ticket. Speaking of confusion, “Perl 6 v6.d” is weird, and other alternatives (“Perl 6.d”, “Perl v6.d”) are even weirder. So one could argue that “Perl 6” is what needs to be retracted, but that's something we want to avoid. Let's discuss that elsewhere if at all.

@jnthn

This comment has been minimized.

Copy link
Member

jnthn commented Jan 21, 2019

I think our overall architecture won't be changing much in the sense that we have the three layers (VM, bootstrapped subset including compiler tools, and full language compiler), and we need to be able to talk clearly about those as we work on them. However, I'd agree that most users won't need to know about anything other than Rakudo. I note that rakudo.org actually does a good job of only talking about Rakudo, and actually so does the perl6.org page. Are there other places that we should deemphasize NQP/MoarVM?

NQP – what are the chances that some other compiler will use NQP

If there were somebody championing that, probably a bit higher than now. If they were, though, they might not find "NQP" the most useful name for marketing it. We also have the situation now where "NQP" refers to both the language and the compiler toolkit, but given we apparently have more names for things than might be ideal, perhaps that's not a problem we want to solve. ;)

So what exactly is the problem that we are trying to solve with Rakudo Star? Is it packaging for Windows? If so, what if we call it “Rakudo for Windows”?

That's something I've also wondered. I think that providing Windows binaries is a really important thing, and that they need to include, at a minimum, zef, so one can install other things.

There's long been the idea that there can then be different distributions, which really are sets of modules, tools, etc. that ship with a compiler. Those could probably be done in suitable ways for different platforms (on Windows just a download of all the things, on Linux a package that depends on the Rakudo package plus the various module packages, etc.) I think there's value in curated sets of modules and tools, but they really need somebody with a vision/goal for them.

Speaking of confusion, “Perl 6 v6.d” is weird, and other alternatives (“Perl 6.d”, “Perl v6.d”) are even weirder

Personally I don't find Perl 6.d particularly weird, if wanting to refer to a specific language version; I'm pretty sure that's what I've been saying out loud when doing talks. Having a codename allows for saying "Perl 6 Diwali" instead, though - while I do like the codenames - I can't say I've been using them it much. On the one hand, picking a holiday also gives some natural theme to the marketing, though I guess there's also always a risk of accidentally offending those who celebrate said holiday. Maybe sometimes it's better to be boring. :-)

@pmurias

This comment has been minimized.

Copy link
Contributor

pmurias commented Jan 22, 2019

If there were somebody championing that, probably a bit higher than now. If they were, though, they might not find "NQP" the most useful name for marketing it. We also have the situation now where "NQP" refers to both the language and the compiler toolkit, but given we apparently have more names for things than might be ideal, perhaps that's not a problem we want to solve. ;)

I'm under the impression that this mix up creates a lot of confusion to new people. I remember people thinking that we compile things down to textual NQP or that you could port rakudo by reimplementing NQP the language.

@FCO

This comment has been minimized.

Copy link
Member

FCO commented Jan 24, 2019

@lizmat

This comment has been minimized.

Copy link
Contributor

lizmat commented Jan 24, 2019

Datapoint from the real world wrt name: https://news.ycombinator.com/item?id=18993064

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.