| 1 |
On December 29, 2020 4:39:06 AM EST, "Michał Górny" <mgorny@g.o> wrote: |
| 2 |
>On Mon, 2020-12-28 at 23:18 +0000, Peter Stuge wrote: |
| 3 |
>> Michał Górny wrote: |
| 4 |
>> > > A. It is a distinct implementation with probably /quite some/ |
| 5 |
>> > > stable |
| 6 |
>> > > compatibility, meaning that it will work perfectly fine as an |
| 7 |
>> > > alternative in many cases. |
| 8 |
>> > |
| 9 |
>> > Except that it doesn't, as has been proven numerous times. |
| 10 |
>> |
| 11 |
>> I'm sure that there are numerous cases where libressl doesn't work, |
| 12 |
>> but that's no reason to dismiss cases where it *does*. |
| 13 |
> |
| 14 |
>Are you asking people to put an effort into maintaining something that |
| 15 |
>can't be practically installed? 'I don't use a single Qt application |
| 16 |
>(yet!), so I demand you support LibreSSL for me!' |
| 17 |
> |
| 18 |
|
| 19 |
I don't see Peter demanding anything here. He has an opinion as a user and he has offered it. Just as others have offered theirs. |
| 20 |
|
| 21 |
>A side effect of this approach is that users will be tricked into using |
| 22 |
>LibreSSL, only to discover that they eventually have to transition |
| 23 |
>their systems back. We have been there with LibAV. However, unlike |
| 24 |
>LibAV, transition between SSL providers is non-trivial. |
| 25 |
> |
| 26 |
|
| 27 |
No one is tricking anyone into using LibreSSL. |
| 28 |
|
| 29 |
>> Did anyone gather actual numbers? |
| 30 |
> |
| 31 |
>$ find -name '*libressl*.patch' | grep -v dev-libs/libressl | wc -l |
| 32 |
>61 |
| 33 |
> |
| 34 |
>That's just the tip of the iceberg. Nobody's going to be able to even |
| 35 |
>guess how many upstream have accepted *bad* patches. How many packages |
| 36 |
>are forcing obsolete algorithms because someone added '!libressl' |
| 37 |
>conditions because LibreSSL keeps pretending to be a newer OpenSSL |
| 38 |
>version without actually implementing its API. |
| 39 |
> |
| 40 |
|
| 41 |
That's quite an assumption to assume all of these patches are "bad." |
| 42 |
|
| 43 |
>> > > B. It brings its own TLS API, a unique feature which by itself |
| 44 |
>> > > warrants the package. |
| 45 |
>> > |
| 46 |
>> > ...which by itself has no future |
| 47 |
>> |
| 48 |
>> That's arrogant and silly coming from anywhere but upstream. |
| 49 |
>> |
| 50 |
>> You can argue that you will never use the API in your TLS programs, |
| 51 |
>> but even then that says really nothing about the API provider itself. |
| 52 |
> |
| 53 |
>That's my opinion and I have a right to have it without being insulted. |
| 54 |
> |
| 55 |
|
| 56 |
Yes, you do. Just as others have a right to theirs without being spoken to as you have done so here. |
| 57 |
|
| 58 |
It is clear you have a problem with the LibreSSL implementation, but portraying that as a user being ignorant, demanding, or anythimg else is uncalled for to get your point across. |
| 59 |
|
| 60 |
Quit saying people are insulting you when you have clearly done so in the very same email response. |
| 61 |
|
| 62 |
>I don't dispute whether it's good. I dispute whether tightly binding |
| 63 |
>it to a problematic LibreSSL is a good idea. Sure, this means that |
| 64 |
>some people will be forced to use LibreSSL because of libtls. |
| 65 |
> |
| 66 |
>But in practice, this means that any upstream starting to use libtls |
| 67 |
>does a huge disservice to their users by preventing them from using |
| 68 |
>their preferred SSL provider. So in the end, you either don't use |
| 69 |
>libtls or your users are in pain. |
| 70 |
> |
| 71 |
>Actually, I see that someone has apparently forked libtls into libretls |
| 72 |
>(the irony!) that works against OpenSSL [1]. |
| 73 |
> |
| 74 |
>[1] https://git.causal.agency/libretls/about/ |
| 75 |
|
| 76 |
Anyway, +1 for LibreSSL removal. |
| 77 |
|
| 78 |
-- |
| 79 |
Sent from my Android device with K-9 Mail. Please excuse my brevity. |
Archives