| View previous topic :: View next topic |
| Author |
Message |
Ant P.
Watchman
Joined: 18 Apr 2009
Posts: 6920
|
 Posted: Mon Aug 04, 2014 2:11 pm Post subject: |
 |
|
| repoman is supposed to do that. Try running it over some large overlays and guess if anyone bothers to use it... |
|
| Back to top |
|
 |
desultory
Bodhisattva
Joined: 04 Nov 2005
Posts: 9410
|
| Posted: Tue Aug 05, 2014 3:59 am Post subject: |
|
|
| hasufell wrote: | | Yes, it's better if that codebase dies, but as you can see we have too many people who are too stubborn to accept that, so it will not happen. |
Perhaps "death by refactoring". Given that there is essentially a newly formed team taking over the work on portage, it seems reasonable to take some time to see what they can do with the code and specifications that they have to work with before throwing out their work and that of those who came before them.
| hasufell wrote: | | * emotionally attached to portage |
Using that as an argument in favor of Paludis is risible, at best, considering the frequently aggressive, even sycophantic, nature of the "support" Paludis often receives. In short, its purported supporters are often its biggest drawback.
This is, obviously, a social problem but that does not somehow exclude it from the set of problems which would need to be addressed before considering transitioning to another package manager as the official or default selection. While Paludis does receive well mannered and well reasoned support it does still have, in some regards, a stigma to overcome. Something which inflammatory pronouncements will not help with.
| hasufell wrote: | | * "works for me" argument, although the bug reports keep piling up and ugly user workarounds have become the default |
Curiously, "it works for me" is the very essence of your argument for using Paludis; and what constitutes an "ugly user workaround" in this context is largely a matter of opinion. Especially when bereft of examples.
| hasufell wrote: | | * most people who like portage have never looked at the code |
Which is markedly distinct from most other software, including Paludis, in exactly what way? Really, if you are going to argue that the code needs reworked, make that argument, not "users are not reading it so it must be bad" which is in itself meaningless in regards to the vast majority of software.
| hasufell wrote: | | * not-invented-here syndrome when it comes to paludis |
Do enlighten us, how exactly did Paludis come into being? Might it have had something to do with a tool known as Qualudis? Might that tool have been developed by one or more individuals who were at the time Gentoo developers for use in their work on Gentoo?
| hasufell wrote: | | * "paludis was written by Ciaran, so we cannot use it" (err) |
A fatuous argument, given the other things which he wrote that are still in use in the tree and elsewhere. Though they are generally heavily patched in some way by now, given the amount of time which has passed since he had access to the respective repositories.
| hasufell wrote: | | * "portage is the fastest PM" (just wrong) |
It might be an incomplete statement, but it is not "just wrong". Again, make the argument as to how one is actually faster in a given use case instead of making what may as well be a blind claim in favor of your own preference. Run a few test cases, time them, publish your results and the configurations and arguments used for each command. In the long run it will take less time than merely repeating that one or the other is faster or slower or more or less reliable. |
|
| Back to top |
|
|
steveL
Watchman
Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery
|
| Posted: Tue Aug 05, 2014 1:32 pm Post subject: |
|
|
| hasufell wrote: | | But your repeated efforts to come after me personally is pretty confusing. If you have a problem, contact me privately. |
I haven't come after you personally, for the record.
| Quote: | | In contrary to a lot of gentoo devs, I answer even private mails and I have never had a problem to say "sorry" in case I messed up in one or another way. |
Yet you refuse even the slightest correction about ignoring someone else's points.. Note not my point, I merely pointed out that you were missing the point being made to you by krinn. And then I got a load of flak for it; along with a later: "I used the word 'social', so I've already covered it," which is part of what I found so amusing.
You're right: it's impossible to change things with "developers" like that. That's the problem with vicious, nasty discussions of the type engendered by paludroids, following their "great leader," even when they don't like it themselves. At that point one hopes they might start to ask more fundamental questions (such as: "Why are we even behaving like this? It's not even fun.") but more often they do what you've done in this thread, which is tedious and boring for anyone who grew past that type of behaviour by 12 (and girls get there a lot sooner.)
Though those types of discussion made good practice for dealing with systemd-fanbois, so it's all grist to the mill, afaic.
Forgive me for not wanting such crap to pollute my inbox: keep the discussion to the forum or the mailing-list. I for one have zero interest in private correspondence with you. Same as several of your fellows whom I've had to tell to stop bothering me off-list. IDK why you guys think it's something of a privilege to interact with you: as you've pointed out yourself, collectively you're a bunch of vipers. Sorry, but my time is worth much more than that, and an hour I spend sorting out one of your heads, is an hour I don't get paid for.
There's no return either, since in every case I've dealt with, you guys come out with similarly tail-chasing argumentation. Reading you decide that you don't have anything to worry about, it's all "some other group" that needs to change (yet again) is simply boring at this point. Grow up, don't grow up, makes no odds to me: spending that time on your latest amateur to start to smell the coffee one-to-one, won't improve the distro, that's for sure, so decision made: keep the self-delusional cruft out of my mailbox.
--
If you wish, ignore the second part of this post: only the first part really is of concern to me since so many of your fellow devs have made a habit of spinning out unanswered criticism of that sort into a cause of bs CoC action. The rest is in the likely vain hope that you'll start to see the light, but if you want to go into more adenoid criticism, don't expect an answer.
Except the occasional LMAO, which is likely the only part that will actually register. |
|
| Back to top |
|
|
Navar
Guru
Joined: 20 Aug 2012
Posts: 353
|
| Posted: Wed Aug 13, 2014 2:55 am Post subject: |
|
|
Succinct and on topic as always.
If something isn't personal, then don't make it such. Cut out the repeated bits of vitriol over and over in various postings regarding age 12 fixation leading to childish assertions. It's simply name calling, trite, tiresome, weird, and entirely disrespecting whomever you're directing it at this time. It's also utilized everywhere in politics as a similar smear tactic. Utilizing hyperbolic comparisons to someone/something else (where do people think Godwin's law came from). Using 'you' primarily in a direct and confrontational manner, and, in general, attempting to act holier-than-thou.
I was looking forward to a reasonable response to on topic questions after a week and instead I find more of this nonsense.
Can we please bring this back on topic? Or if this is the manner in which such a thread should de-rail, then politely moderate (moving unrelated inflamatory posts out) and/or lock it down as this is going the same directions as others. Thanks.
Last edited by Navar on Wed Aug 13, 2014 3:04 am; edited 1 time in total |
|
| Back to top |
|
|
Navar
Guru
Joined: 20 Aug 2012
Posts: 353
|
| Posted: Wed Aug 13, 2014 3:01 am Post subject: |
|
|
| Krinn wrote: | | And no dev ever think creating an "ebuild-validator" that check ebuild and issue warning/hint/errors that might resolve most of them before they are even push in the tree? |
| Ant P. wrote: | | repoman is supposed to do that. Try running it over some large overlays and guess if anyone bothers to use it... |
And to use it is documented. My impression in use and the documentation was left that of being a policy enforcer on QA than ebuild style, etc. Which is why I asked for some real details regarding:
| hasufell wrote: | | [...] bad practices in ebuild writing that people think are good. |
|
|
| Back to top |
|
|
steveL
Watchman
Joined: 13 Sep 2006
Posts: 5153
Location: The Peanut Gallery
|
| Posted: Thu Aug 14, 2014 2:23 pm Post subject: |
|
|
| Pfft the vitriol is coming from you. |
|
| Back to top |
|
|
TomWij
Retired Dev
Joined: 04 Jul 2012
Posts: 1553
|
| Posted: Thu Aug 14, 2014 5:13 pm Post subject: |
|
|
| Navar wrote: | | Let's start there. From an outsider looking in, is there some concise and clear references/thread anything somewhere that refutes any issues showing in the existing published developer docs for considered best practices? I'd prefer to avoid a 300 page set of side rants or anything that simply strays away from the technical facts towards personal attacks. |
There is Common Mistakes; a lot of other known bad practices are spread out over Bugzilla, MLs, Wiki and other communication mediums. Having a vocal minority complain about them is quite easy, given that you can just state it and be done with it; however, getting everyone's attention through knowledge codification by formally listing them is much harder as you need others to agree with you that it is a bad practice. Other than that one needs to spend the resources on getting this to happen.
| Navar wrote: | | My impression in use and the documentation was left that of being a policy enforcer on QA than ebuild style, etc. |
Repoman is somewhat a mixture of ebuild and workflow mistakes, as well as policy enforcement by various Gentoo Projects (eg. G2CONF deprecation by GNOME team, ...) including QA, ... |
|
| Back to top |
|
|
krinn
Watchman
Joined: 02 May 2003
Posts: 7470
|
| Posted: Fri Aug 15, 2014 2:53 am Post subject: |
|
|
| TomWij wrote: | | Having a vocal minority complain about them is quite easy, given that you can just state it and be done with it; however, getting everyone's attention through knowledge codification by formally listing them is much harder as you need others to agree with you that it is a bad practice. |
You can have the repo auto-run the "ebuild validator" and just drop any ebuild that doesn't pass the tool tests : no need to look who made it and where it came from, if it try enter the tree ; it must comply with the rules.
I'm unsure how anyone would called that, but that must be close to what should be called "QA work". |
|
| Back to top |
|
|
shazeal
Apprentice
Joined: 03 May 2006
Posts: 206
Location: New Zealand
|
| Posted: Fri Aug 15, 2014 4:43 am Post subject: |
|
|
I was intrigued by this Paludis malarkey, so I installed it. Synced it up. Then I ran into this...
http://paludis.exherbo.org/trac/ticket/817
And I happily noped the hell out of there. Can you can explain to me how I can use real multilib on amd64 with Paludis without having to learn how to write a profile to do so? 
_________________
CFLAGS="-OmgWTFR1CE --fun-lol-loops --march=asmx86go" |
|
| Back to top |
|
|
TomWij
Retired Dev
Joined: 04 Jul 2012
Posts: 1553
|
| Posted: Fri Aug 15, 2014 9:57 am Post subject: |
|
|
| krinn wrote: | | You can have the repo auto-run the "ebuild validator" and just drop any ebuild that doesn't pass the tool tests : no need to look who made it and where it came from, if it try enter the tree ; it must comply with the rules. I'm unsure how anyone would called that, but that must be close to what should be called "QA work". |
Committing to the Portage tree requires the use of Repoman; Repoman doesn't allow you to commit known errors, unless you pass along an explicit --force parameter which usage will be mentioned in the commit message. Those who do not use Repoman or deliberately use --force in a way that attempts to break those rules that are set up as errors get their commit access removed.
Apart from that commit barrier, Patrick runs repoman over the whole Portage tree every few hours; which differences are output to #gentoo-bugs and #gentoo-qa on IRC, as well as are visible on http://packages.gentooexperimental.org/repoman-current-issues.txt and in category form. More about how this works can be read about in his blog post. |
|
| Back to top |
|
|
krinn
Watchman
Joined: 02 May 2003
Posts: 7470
|
| Posted: Fri Aug 15, 2014 12:39 pm Post subject: |
|
|
That's pure example but
app-backup/rdiff-backup: DESCRIPTION is 106 characters (max 100)
If you don't force repoman to reject all ebuild that fail, you then MUST go into a millions line of text speaking with the dev that does that to explain him blahbhlabhalhbah...
And i'm sure the scenario is not sci-fi if i say : devs goes into dev-ml and complain someone shouldn't do that or this ; other devs doing the same backup the "faulty" dev... blhablhalbhabhlhabahl 10 pages later ; nothing change, except devs have words with each other for no result.
Now if you add to that you must do the same with every devs that are in fault...
Just force repoman to simply reject it with that error and you'll see the same dev will magically recommit an ebuild limit to 100 chars !
Else you get something like that :
| TomWij wrote: | | Having a vocal minority complain about them is quite easy, given that you can just state it and be done with it; however, getting everyone's attention through knowledge codification by formally listing them is much harder as you need others to agree with you that it is a bad practice. Other than that one needs to spend the resources on getting this to happen. |
Sure, some rules seems bad and may need a change (like the too big patch one that looks kinda stupid ; ok for me).
But ENFORCE THEM EVEN THE STUPID ONES.
So people will need to discuss them and change them instead of ignoring them.
Because as soon as one rule ; even the biggest stupid rule that exist, is ignore : all others rules "not that stupid" can be also ignore, as any rule will be seen as stupid by someone hit by it, so all rules are stupid -> all rules are ignore. |
|
| Back to top |
|
|
Ant P.
Watchman
Joined: 18 Apr 2009
Posts: 6920
|
| Posted: Sat Aug 16, 2014 4:08 pm Post subject: |
|
|
| shazeal wrote: | | Can you can explain to me how I can use real multilib on amd64 with Paludis without having to learn how to write a profile to do so? |
How is the default/linux/amd64/13.0 profile broken? Am actually interested in knowing; I've been using no-multilib but thinking of changing, but if the QA is as bad as it sounds... |
|
| Back to top |
|
|
TomWij
Retired Dev
Joined: 04 Jul 2012
Posts: 1553
|
| Posted: Sat Aug 16, 2014 7:31 pm Post subject: |
|
|
| krinn wrote: | That's pure example but
app-backup/rdiff-backup: DESCRIPTION is 106 characters (max 100) |
Word in advance; we don't turn the world upside down for a few extra characters, yet we warn about it so people know to keep it short where possible (for proper output purposes).
| krinn wrote: | | If you don't force repoman to reject all ebuild that fail, you then MUST go into a millions line of text speaking with the dev that does that to explain him blahbhlabhalhbah... |
There are developers that want these warnings to be forced to act as errors, eg. Patrick; they however have not yet succeeded in convincing the Gentoo Council to make this change succeed, whom have discussed this Gentoo Council agenda item before. Given that other developers complain that they don't want to do the extra efforts for these unimportant warnings, such force might never happen. Meanwhile these warnings can be fixed by the QA team without necessarily talking to the developer; we do this where this is possible without needing the developer's interaction, otherwise it would take ages to fix anything at all.
| krinn wrote: | And i'm sure the scenario is not sci-fi if i say : devs goes into dev-ml and complain someone shouldn't do that or this ; other devs doing the same backup the "faulty" dev... blhablhalbhabhlhabahl 10 pages later ; nothing change, except devs have words with each other for no result.
Now if you add to that you must do the same with every devs that are in fault... |
Commit mistake discussions happens, but not so often. Pointing out another developer's commit mistakes on the ML in public isn't so acceptable as doing it in private, unless it makes sense for the rest of the community to follow a better practice and learn from it; in this case we can try to translate it to a warning or error in Repoman. Apart from that; some of these discussons indeed result in no final decision, warning or error. Such discussions however make clear how the developer community thinks about it; that is, it clarifies that either ebuild style is good (unless consistency is needed, then QA or Council votes on it). |
|
| Back to top |
|
|
shazeal
Apprentice
Joined: 03 May 2006
Posts: 206
Location: New Zealand
|
| Posted: Sat Aug 16, 2014 9:24 pm Post subject: |
|
|
| Ant P. wrote: | | shazeal wrote: | | Can you can explain to me how I can use real multilib on amd64 with Paludis without having to learn how to write a profile to do so? |
How is the default/linux/amd64/13.0 profile broken? Am actually interested in knowing; I've been using no-multilib but thinking of changing, but if the QA is as bad as it sounds... |
You cannot un-mask use flags using Paludis... So for example to use Multilib on amd64 using emerge you just do this...
/etc/portage/profile/use.mask
Using Paludis you write a profile that unmasks the abi_x86_32 use flag. This requires understanding how to write one, and I would assume keep it updated, this leads to more complexity and probably a much larger chance of shafting your system.
You will get zero support from the Paludis developers doing so since doing so means you are some kind of insane lunatic according to that support thread. Paludis is the North Korea of package managers IMO 
_________________
CFLAGS="-OmgWTFR1CE --fun-lol-loops --march=asmx86go" |
|
| Back to top |
|
|
hasufell
Retired Dev
Joined: 29 Oct 2011
Posts: 429
|
| Posted: Tue Aug 19, 2014 2:40 am Post subject: |
|
|
| shazeal wrote: | | Ant P. wrote: | | shazeal wrote: | | Can you can explain to me how I can use real multilib on amd64 with Paludis without having to learn how to write a profile to do so? |
How is the default/linux/amd64/13.0 profile broken? Am actually interested in knowing; I've been using no-multilib but thinking of changing, but if the QA is as bad as it sounds... |
You cannot un-mask use flags using Paludis... So for example to use Multilib on amd64 using emerge you just do this...
/etc/portage/profile/use.mask
|
Erm?
Unmasking use flags using paludis is totally equivalent to portage except that you have to add an eapi file with only "5" in it to the profiles/ folder.
The fact that portage does not need this file is not relieving. |
|
| Back to top |
|
|
hasufell
Retired Dev
Joined: 29 Oct 2011
Posts: 429
|
| Posted: Tue Aug 19, 2014 3:01 am Post subject: |
|
|
| Navar wrote: | | hasufell wrote: | | [...] bad practices in ebuild writing that people think are good. |
|
recent details involve relying on a feature which is
* undocumented in PMS
* buggy/broken/misdesigned
* PM-specific
* causing trouble for everyone who either disables it or uses a different PM
* slowing down portage
But that is a topic for its own and has been discussed on the gentoo dev ML. It was about dynamic deps.
And we have more of it, like overuse of REQUIRED_USE which was _never_ meant as a common way to express dependencies between configure switches. It's only purpose was to not make reverse deps break. Still people use it in all sorts of random ebuilds and force users to micro-manage flags in an incredibly useless way, as well as making tinderboxing harder. It also causes unreadable error messages in package managers and confuses them in weird ways, breaking more features like portage suggestions on how to fix stuff.
But again, this is really offtopic. |
|
| Back to top |
|
|
hasufell
Retired Dev
Joined: 29 Oct 2011
Posts: 429
|
| Posted: Tue Aug 19, 2014 3:23 am Post subject: |
|
|
| desultory wrote: | | hasufell wrote: | | Yes, it's better if that codebase dies, but as you can see we have too many people who are too stubborn to accept that, so it will not happen. |
Perhaps "death by refactoring". Given that there is essentially a newly formed team taking over the work on portage, it seems reasonable to take some time to see what they can do with the code and specifications that they have to work with before throwing out their work and that of those who came before them. |
Afaik they are essentially just trying to keep portage alive until alternatives are back on track, like pkgcore.
No one intends to rewrite portage. The only one who claimed that is TomWij, but I haven't seen any code yet.
| desultory wrote: | | hasufell wrote: | | * "works for me" argument, although the bug reports keep piling up and ugly user workarounds have become the default |
Curiously, "it works for me" is the very essence of your argument for using Paludis; and what constitutes an "ugly user workaround" in this context is largely a matter of opinion. Especially when bereft of examples. |
I think there were multiple plain technical arguments in this thread, including superior features, better dependency resolution, higher strictness and more control over the dependency resolution options. I will talk my mouth dry if I start commenting on every single one of them, so I'd rather point you to the documentation/code.
And I already mentioned some downsides of paludis as well, including poor support (on both sides of the gates).
| desultory wrote: | | hasufell wrote: | | * most people who like portage have never looked at the code |
Which is markedly distinct from most other software, including Paludis, in exactly what way? Really, if you are going to argue that the code needs reworked, make that argument, not "users are not reading it so it must be bad" which is in itself meaningless in regards to the vast majority of software. |
The point was that even some of the current portage developers are sick of looking at portage code. Hacking can be fun, though, but it apparently is not in this case.
| desultory wrote: | | hasufell wrote: | | * "portage is the fastest PM" (just wrong) |
It might be an incomplete statement, but it is not "just wrong". Again, make the argument as to how one is actually faster in a given use case instead of making what may as well be a blind claim in favor of your own preference. Run a few test cases, time them, publish your results and the configurations and arguments used for each command. In the long run it will take less time than merely repeating that one or the other is faster or slower or more or less reliable. |
Not sure why I have to waste my time setting up scientific test cases to disprove incomplete statements. I'd expect those guys who claimed portage is faster to do that.
They didn't even bother to test the different dependency resolution options like "-z".
If you insist, I'll use the term "without any real evidence" instead of "wrong". |
|
| Back to top |
|
|
hasufell
Retired Dev
Joined: 29 Oct 2011
Posts: 429
|
| Posted: Tue Aug 19, 2014 3:31 am Post subject: |
|
|
| Navar wrote: | | Can we please bring this back on topic? |
You are right. I'll ignore those non-technical posts from now on. Don't hesitate to report them. |
|
| Back to top |
|
|
desultory
Bodhisattva
Joined: 04 Nov 2005
Posts: 9410
|
| Posted: Tue Aug 19, 2014 3:57 am Post subject: |
|
|
| hasufell wrote: | | Afaik they are essentially just trying to keep portage alive until alternatives are back on track, like pkgcore. |
At least when the team initially formed, I had the distinct impression that they intended to take up regular maintenance.
| hasufell wrote: | | No one intends to rewrite portage. The only one who claimed that is TomWij, but I haven't seen any code yet. |
I had thought there were others who had made some mention of refactoring or forking portage, outside of the official project, without any visible effect, though I could not reliably put names to any others at the moment.
| hasufell wrote: | | If you insist, I'll use the term "without any real evidence" instead of "wrong". |
It would be a welcome change from the tendency toward rhetorical stances which end up leaving what would be valid arguments lingering as readily dismissed technically wrong claims. Even so, I had meant to offer suggestions for making better arguments, not to insist upon a particular imposed vernacular. |
|
| Back to top |
|
|
shazeal
Apprentice
Joined: 03 May 2006
Posts: 206
Location: New Zealand
|
| Posted: Tue Aug 19, 2014 8:03 pm Post subject: |
|
|
| hasufell wrote: | Erm?
Unmasking use flags using paludis is totally equivalent to portage except that you have to add an eapi file with only "5" in it to the profiles/ folder.
The fact that portage does not need this file is not relieving. |
I dont understand, none of this is documented afaik. Can you explain exactly how you would go about doing this? That support post suggests that you need to write a profile for each repository which unmasks the use flag.
You are saying I just create a file called eapi with 5 in it and it unmasks stuff?
Another thing that seems overly complicated is the package.provided, I dont want nor use the gentoo-sources, I make my own kernel. The documentation simply states...
| Quote: |
Portage supports pretending that a package is installed for the purposes of dependency resolution via a file called package.provided. Paludis provides a cleaner approach, in the form of cave import. cave import allows you to more cleanly reinstall, uninstall, and upgrade unpackaged packages, using the same sort of merge/unmerge process that paludis uses for ebuilds.
To use cave import, you will first have to configure an installed_unpackaged format repository.
To get a more direct equivalent of package.provided, you can "install" an empty directory with cave import. |
What does this even mean? Seriously, it just feels so damn over complicated. There seem to be zero examples of how to do this kind of thing anywhere in the documentation unless the official site is the wrong place to look?
Also this... What the actual f?
| Quote: |
Sync while doing anything else
If you think this'll work, buy a mac. |
Trolling in the docs?
The site claims Paludis to be stricter, and constantly bashes on the broken state of emerge. Yet it feels like Ive gone back 20 years and am dialing up some BBS and trying to ask for help from some socially retarded sysop when I try and read the documentation.
Paludis seems like a nice idea, and when I read the features it sounds amazing, like its going to rock my world. But as soon as I try and start using it I feel like I am a slave to the system everything takes so much work to get anywhere. And that is saying alot considering how much work emerge takes compared with other distros package mangers.
_________________
CFLAGS="-OmgWTFR1CE --fun-lol-loops --march=asmx86go" |
|
| Back to top |
|
|
Navar
Guru
Joined: 20 Aug 2012
Posts: 353
|
| Posted: Sat Aug 23, 2014 4:35 am Post subject: |
|
|
| hasufell wrote: | | recent details involve relying on a feature which is [...] |
Hm. Strong enough points. Why isn't repoman being used to prevent these issues in particular?
| hasufell wrote: | | But that is a topic for its own and has been discussed on the gentoo dev ML. It was about dynamic deps. |
That thread I have been reading through off and on when I have time. I had gotten the impression earlier that sub slotting was also causing headaches.
| hasufell wrote: | | And we have more of it, like overuse of REQUIRED_USE which was _never_ meant as a common way to express dependencies between configure switches. It's only purpose was to not make reverse deps break. Still people use it in all sorts of random ebuilds and force users to micro-manage flags in an incredibly useless way, as well as making tinderboxing harder. It also causes unreadable error messages in package managers and confuses them in weird ways, breaking more features like portage suggestions on how to fix stuff. |
This I have quietly questioned a number of times; whether the restrictions were valid or not and how often non-obvious to resolve it is for users. I can only imagine the pain and interruptions of multi profile/use tinderbox combinations. I've just dealt with the issues as they've affected me passively thinking there was a direct cause/need for their use. I glance at the dev ML archives when I find time or a particular topic I'm searching on. Is there any effort/reason why there's not say a monthly list (by package/ebuild revision) trying to resolve overuse/abuse of that particular matter? Like in the way I've seen lists generated there before for maintainer-ship (temp proxy or abandoned) or tree removal. |
|
| Back to top |
|
|
Navar
Guru
Joined: 20 Aug 2012
Posts: 353
|
| Posted: Sat Aug 23, 2014 4:56 am Post subject: |
|
|
| hasufell wrote: | | I'll ignore those non-technical posts from now on. Don't hesitate to report them. |
No need. I wasted enough time with regards to. Besides, you have a site-admin/mod posting on the thread and they generally do view interesting/multi-page threads. I'm left with the impression they prefer mostly community policing. The particular individual I objected to and myself apparently don't see eye-to-eye as of late and that's fine; opinions be what they may. |
|
| Back to top |
|
|
Navar
Guru
Joined: 20 Aug 2012
Posts: 353
|
| Posted: Sat Aug 23, 2014 5:46 am Post subject: |
|
|
| shazeal wrote: | Trolling in the docs?
The site claims Paludis to be stricter, and constantly bashes on the broken state of emerge. Yet it feels like Ive gone back 20 years and am dialing up some BBS and trying to ask for help from some socially retarded sysop when I try and read the documentation. |
Hey, some things 20 years ago were better. Particularly documentation because money actually used to be applied towards that.
Yeah that is unfortunate. I've glanced at it briefly a number of times, but there was a lot of flack posts against here on the forums in previous years to keep me wary of looking into further (including claims of abandoned). It's been about 8 years now? I think it's NIH syndrome, bike shedding and just good ol' competitive we disagree with them approach. And that's fine too, but whatever happened to a wink, nod and evil smile instead for the coming race?
If they end up producing something that performs consistently better (reliable, accurate, performance, capable), it'll eventually help supersede portage on merit, if portage falls by the wayside. Even better if a core engine part can be modularized away with respective API so any front end in any language could leverage it (thereby essentially reproducing the look/feel of existing portage or whatever flavorings others preferred). I consider portage to be such an attempt and 'emerge' as just a CLI wrapper to leverage.
I've written primarily in C/C++ for several decades. It would greatly surprise me, given like algorithms, etc. that an interpreted language, including bytecode 'compiled' VM intermediates (particularly with auto GC), would perform better than a system and compiled language. This is on the presumption that well performing and correct code was wrote in each to begin with. Plus there is the whole bash environment support that I'm not even adding for further complicating this. And how much of that do we need to retain?
Maybe what needs to be done to portage is a tightened restriction on clear documentation, API, and some flexibility removed to eliminate overly complicated code to remove excessive outer range edge cases. I recall seeing some profiling commentary from TomWij in the past regarding the portage slowdowns, but I'm unsure how much further that has went. |
|
| Back to top |
|
|
apathetic
n00b
Joined: 28 Aug 2014
Posts: 36
|
| Posted: Thu Aug 28, 2014 2:11 pm Post subject: |
|
|
| hasufell wrote: | | You probably missed the spot where I said that gentoo is socially broken. |
This. |
|
| Back to top |
|
|
shazeal
Apprentice
Joined: 03 May 2006
Posts: 206
Location: New Zealand
|
| Posted: Thu Aug 28, 2014 7:26 pm Post subject: |
|
|
| Navar wrote: | | If they end up producing something that performs consistently better (reliable, accurate, performance, capable), it'll eventually help supersede portage on merit, if portage falls by the wayside. Even better if a core engine part can be modularized away with respective API so any front end in any language could leverage it (thereby essentially reproducing the look/feel of existing portage or whatever flavorings others preferred). I consider portage to be such an attempt and 'emerge' as just a CLI wrapper to leverage. |
Without adoption it wont gain any traction. Without documentation it wont gain any adoption. Unless as per Portage it is forced on users. I honestly had not looked at the Portage documentation in years, I dont even remember the last time I looked. So I decided to have a look...
It is worse than the Paludis docs (Except for the lack of snide comments throughout). So I guess that explains why whoever wrote the Paludis docs aimed so low. The documentation for Portage is sparse and hidden, I had to use Google to find most information.
So I decided to see what documentation there is on unmasking a use flag in Portage, so I could compare it with Paludis... And here is the "documentation" on how to do that...
https://forums.gentoo.org/viewtopic-t-376608.html
So I would add one more thing to hasufell's list of reason people still use Portage.
* Been using Portage for so long they didn't realize how rubbish the documentation is.
_________________
CFLAGS="-OmgWTFR1CE --fun-lol-loops --march=asmx86go" |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Gentoo Chat
