| 1 |
On July 8, 2021 9:46:11 PM UTC, Matt Turner <mattst88@g.o> wrote: |
| 2 |
>On Thu, Jul 8, 2021 at 2:34 PM MSavoritias |
| 3 |
><marinus.savoritias@×××××××.dev> wrote: |
| 4 |
>> On July 8, 2021 8:50:39 PM UTC, Matt Turner <mattst88@g.o> |
| 5 |
>wrote: |
| 6 |
>> >On Thu, Jul 8, 2021 at 1:41 PM Peter Stuge <peter@×××××.se> wrote: |
| 7 |
>> >> |
| 8 |
>> >> Matt Turner wrote: |
| 9 |
>> >> > If you can find a case where you wouldn't want to enable one of |
| 10 |
>> >these |
| 11 |
>> >> > USE flags, please let me know and I'll reconsider my position. |
| 12 |
>> >> |
| 13 |
>> >> My catalyst spec files all have use: -* foo bar x y z |
| 14 |
>> >> specifically because the defaults are never what I want for any |
| 15 |
>given |
| 16 |
>> >> system. I build desktops, servers, containers, VM appliance images |
| 17 |
>> >and |
| 18 |
>> >> embedded system, and I know what I want in each one. Especially |
| 19 |
>the |
| 20 |
>> >> latter frequently have only very few USE flags set and I want zero |
| 21 |
>> >> extra dependencies. |
| 22 |
>> > |
| 23 |
>> >I think you're making a great argument that you'd be completely |
| 24 |
>> >unaffected by any of the suggestions in this thread. |
| 25 |
>> > |
| 26 |
>> >> I completely agree that the default USEs should rather be reduced, |
| 27 |
>> >> not increased. Isn't this what profile inheritance is for? It |
| 28 |
>would |
| 29 |
>> >> be great if I didn't essentially have to create my own profile |
| 30 |
>when I |
| 31 |
>> >> want a very minimal system. |
| 32 |
>> >> |
| 33 |
>> >> |
| 34 |
>> >> Matt Turner wrote: |
| 35 |
>> >> > I'd claim most of these packages' bzip2/lzma/zstd USE flags |
| 36 |
>should |
| 37 |
>> >> > be removed in favor of statically enabling them |
| 38 |
>> >> |
| 39 |
>> >> That is the direct opposite of Gentoo's single most core value: |
| 40 |
>> >choice |
| 41 |
>> > |
| 42 |
>> >Choice makes sense when there's a legitimate trade-off to be made. |
| 43 |
>> >Choice isn't dogma. |
| 44 |
>> |
| 45 |
>> Well the legitimate trade-off is complexity as stated previously. |
| 46 |
>Gentoo is not supposed to be batteries included. It is supposed to be |
| 47 |
>building blocks for each persons own thing. |
| 48 |
>> |
| 49 |
>> Instead of adding the use flag what would ne more in Gentoo spirit |
| 50 |
>would be to add to handbook a guide for common use flags. |
| 51 |
>> |
| 52 |
>> Plus just because people disagree here with the proposal doesnt mean |
| 53 |
>its dogma. It may be just disagreement. |
| 54 |
> |
| 55 |
>That's not my claim. |
| 56 |
> |
| 57 |
>It's akin to defending what you said by saying "Well, we have free |
| 58 |
>speech so I can say whatever I want!". Of course you can, but that's |
| 59 |
>not the point. You're not defending the substance of the speech. It's |
| 60 |
>a lazy argument. |
| 61 |
> |
| 62 |
>Similarly, people say "Well, Gentoo is about choice!" even when the |
| 63 |
>choice is absolutely meaningless. Of course Gentoo offers a lot of |
| 64 |
>choice, but that's not the point. You're not defending the value of |
| 65 |
>the choice. It's a lazy argument. |
| 66 |
|
| 67 |
As far as your first point goes as it has been stated adding flags increases complexity along with the size of the packages. Plus not everybody wants these flags so it brings more work to these cases too. What are actually the things missing that we need this flag? Is there actually something required or broken from somewhere? |
| 68 |
|
| 69 |
Just because a use flag doesnt break stuff if its enabled doesnt mean it doesnt have its trade offs. Especially system wide |
| 70 |
> |
| 71 |
>It's easy for people who don't respond to bug reports to discount the |
| 72 |
>overhead every configuration knob adds. |
| 73 |
|
| 74 |
Please dont assume stuff like this. It devalues the conversation. |
| 75 |
|
| 76 |
MSavoritias |
Archives