> On Nov 10, 2014, at 7:27 AM, Richard Elling <[email protected]> wrote:
>
> On Nov 9, 2014, at 7:55 AM, Garrett D'Amore via illumos-discuss
> <[email protected]> wrote:
>
>>
>>> On Nov 9, 2014, at 6:07 AM, Richard Elling <[email protected]> wrote:
>>>
>>> On Nov 8, 2014, at 9:15 PM, Garrett D'Amore <[email protected]> wrote:
>>>
>>>> Most of this work is designed to make life better for developers who build
>>>> products on top of illumos, as opposed to adding customer visible features.
>>>
>>> My point exactly. No customer visible features == no value.
>>> -- richard
>>>
>>
>> This line of thinking would prevent the creation of things like compilers,
>> make, etc. if it were allowed to be the driving force.
>
> Your interpretation of my statement is incorrect. Adding a new compiler
> clearly adds value.
What if the build tool (say make or lex) were not “delivered” but only used to
facilitate our own development? Think e.g. a custom rpcgen? I can see
enormous value in some circumstances like this, even though we might never
document or make that feature directly available to 3rd party developers.
>
>>
>> *Enablement* of developers is its own value.
>
> Agree.
Enablement of developers of illumos-gate itself is its own value. (Not just
3rd party developers.) Do you still agree?
>
>>
>> Anyway, as you’ve not contributed code, nor has Coraid, to the illumos code
>> base, the work I’ve been doing costs you nothing and only benefits you. So
>> I really am not all that interested in this “no features no value” view
>> point.
>
> I can only take this as an intentional, personal, and shameful insult. Not
> just directed at me, but at the great
> folks working hard on delivering value every day at Coraid. Stop it. Stop it
> now.
Um. did you not see “to the illumos code base” ? What has Coraid done to
contribute to illumos, ever? This isn’t a jibe at Coraid, they may have
excellent products and talented engineers, but they aren’t a contributing part
of the illumos community.
I stand by my comment above.
>
> I see a regular pattern of behavior:
> 1. declare a prior work a "crime scene," publicly insult the authors,
> call the code "crap," and eject it
I’ve ejected code that is indeed crap. And yes, I’ve written code that is
crap too. This is not intended to be pejorative of the authors, but yes, I
call a spade a spade. If the code is unreadable, unmaintainable, and filled
with latent bugs — then yes, it is crap. Those aren’t the only metics of
crappiness, but they are the ones I’m most likely to use. I’m hardly alone
here, although perhaps I’m too vocal.
The ejection of code from -gate is not just code that is crap though. (It is
sometimes — the audio utilities that were written in C++ was some really ugly
C++ code, and I think I’d apply the crap moniker to it without reservation.)
The SunOS 4.x work isn’t crap, or at least it wasn’t when it produced. It’s
high quality work. But its high quality work nobody *needs* or uses. So then
it becomes dead weight. By the way, I’ll happily admit that I’ve produced
some dead weight code too. The support for Tadpole SPARCLE laptops fell into
that case — and although I was author, I was also happy to remove it. I’ve
declared the same thing for at least one of the audio drivers that supports
hardware nobody has ever used with illumos or even Solaris outside of my own
testing. So I nuked it. (There are probably other audio drivers that can be
removed.)
So removal is *absolutely* not pejorative towards the authors. But everything
has its time and place.
Oh, and btw, everything that “I’ve” ejected from the -gate has been done with
the approval (both explicit and implicit) of the advocates and the
developer-council.
> 2. wonder why nobody wants to contribute upstream
>
> Rather than spending any time with #1, which adds no value, why not spend
> time upstreaming new features
> from the many downstream repos? This would add value.
Feel free to do so. I don’t own illumos-gate, nor control it, nor have I ever
attempted to do so.
- Garrett
-------------------------------------------
illumos-discuss
Archives: https://www.listbox.com/member/archive/182180/=now
RSS Feed: https://www.listbox.com/member/archive/rss/182180/21175430-2e6923be
Modify Your Subscription:
https://www.listbox.com/member/?member_id=21175430&id_secret=21175430-6a77cda4
Powered by Listbox: http://www.listbox.com
