On Tuesday, 1 December 2015 at 13:40:59 UTC, terchestor wrote:
> botched this one. If Walter made this one everybody would threaten to leave D forever and want his head on a spike. But the mistake must be fixed anyway.

You seriously think DUB is D's biggest problem? You don't even need it to make use of D.

On Tuesday, 1 December 2015 at 13:40:59 UTC, terchestor wrote:
> On Tuesday, 1 December 2015 at 04:38:59 UTC, Saurabh Das wrote:
>> On Monday, 30 November 2015 at 22:59:04 UTC, retard wrote:
>>> Just voted at http://www.easypolls.net/poll.html?p=565587f4e4b0b3955a59fb67 - 140 votes, 75% are against SDL. That should count for something? Sonke?
>>
>> As Sonke had pointed out in the other thread, there was a long process before SDL was adopted and the community was kept informed of it all through. I think everybody in this community had plenty of time and enough number of opportunities to object to SDL if they had any issues with it. See http://forum.dlang.org/post/n36gvd$1lia$1@digitalmars.com
>>
>> It's unfair to criticise SDL at this point after it's been implemented when there was ample opportunity to object earlier. It's not a Douglas Adamsesque "notes were kept in a basement and accessible only after flipping a dice" scenario - in this case, requests for comments were posted in this very forum, on GitHub and elsewhere.
>
> So now that the work is done we're stuck with it forever? Even though it solves the wrong problem and we're better off without it? How does that argument even work?
>
>> Given that JSON is always going to be supported, just use JSON if you are so uncomfortable with SDL.
>
> In a fair world the default should be JSON and SDL should be phased out.
>
> I do appreciate that people are nice with Sonke-who clearly botched this one. If Walter made this one everybody would threaten to leave D forever and want his head on a spike. But the mistake must be fixed anyway.

I didn't say that it should or should not be fixed - that is a separate discussion entirely. Personally I don't find SDL so hot, but I'm okay with using it if needed. It's just a config format after all. Why does it evoke such a strong reaction? SDL or JSON both are perfectly fine.

My point is that it's unfair to criticise Sonke now - who has kept the community abreast of the proposed changes and did not get any feedback on this during the during the many calls for comments and suggestions to be now told that 75% of the community doesn't like the change. Where was 75% of the community when the change was in the open for discussion and amendment?

Development doesn't happen in an isolated shell. Timely and thought out feedback and criticism is much required. Sonke sought it and unfortunately received none of that from us in this matter, so the fault is equally ours.

On Monday, 30 November 2015 at 22:59:04 UTC, retard wrote:
> Just voted at http://www.easypolls.net/poll.html?p=565587f4e4b0b3955a59fb67 - 140 votes, 75% are against SDL. That should count for something? Sonke?

"To make a long thread short, the probable take-aways are dub might switch back to defaulting to json, though sdlang will probably not be removed. We need also better communication about decisions in the community to avoid these kind of after-the-fact misconceptions in the future."

http://arsdnet.net/this-week-in-d/nov-29.html

End of.

On 12/01/2015 03:01 PM, Saurabh Das wrote:
> now told that 75% of the community doesn't like the change.

I'm sorry to enter this discussion, but that's simply not what the poll is saying. It should be very obvious that the poll is basically meaningless.

On Tuesday, 1 December 2015 at 03:49:09 UTC, Adam D. Ruppe wrote:
> simpledisplay.d can do what SDL does :P
>
> oh wait this is about the other SDL
>
> well I want to talk about simpledisplay. I've been doing a documentation book in the ddoc with lots of examples:
>
> http://arsdnet.net/arsd/simpledisplay.html
>
> (similar docs are on the way for cgi, dom, minigui, and simpleaudio btw)
>
> I'm so close to being able to replace SDL with just a couple little D files - no more annoying DLLs!

Can you write a small, compact, up to speed tutorial once you've finished simpledisplay (set up, basic examples)? I'd love to test it.

On Tuesday, 1 December 2015 at 14:37:18 UTC, Chris wrote:
> Can you write a small, compact, up to speed tutorial once you've finished simpledisplay (set up, basic examples)? I'd love to test it.

Well, simpledisplay itself isn't done, but it works for many things now (and myself and several other users make use of it now).

The big change is eventually I am going to change the module name, which is a breaking change, but otherwise I don't have major API changes in mind. (A big addition it needs though is Cocoa Mac OS X support again. It had it years ago then it fell away. I want to redo it when the new ObjC integration in D is done.)

But if you copy/paste the examples in the html file you should be able to get started. The code itself is in this collection:

https://github.com/adamdruppe/arsd

simpledisplay.d and color.d are the two files needed. Then you compile by just adding those two files to your build or I have also made a dub thing for it but I haven't really tested this yet:

http://code.dlang.org/packages/arsd-official%3Asimpledisplay/~master


I do need to write more docs and tutorials but you can play a little with the examples there - the event viewer is first in the example list to give you an idea of what input it gives, then the Pong game underneath that gives a fairly complete demo of basic drawing and input. It can also do things like create OpenGL contexts, which I have a demo of here: http://arsdnet.net/dcode/book/chapter_12/10/opengl.d

(The OpenGL bindings are mostly old, glBegin/glEnd stuff, because that's what I know [!!], but the same context creation works for new functions too so I'm adding that. Just commited a big batch of new stuff yesterday too.)


The audio and joystick parts are in my github repo... and basically work, but I'm still not completely happy with them yet.

On 12/01/2015 09:18 AM, Timon Gehr wrote:
> On 12/01/2015 03:01 PM, Saurabh Das wrote:
>> now told that 75% of the community doesn't like the change.
>
> I'm sorry to enter this discussion, but that's simply not what the poll
> is saying. It should be very obvious that the poll is basically
> meaningless.

Independent on the topic at hand - wondering what your reasoning is. I just took a look and there are 205 votes. Not a large number, but quite a lot more than any voting we saw in the past (when consensus was proclaimed after like 15 votes :o)). Intuitively I agree with you, but I wonder at what point numbers become large enough to capture meaning. Thx! -- Andrei

On Tue, Dec 01, 2015 at 01:43:41PM +0000, Ola Fosheim Grøstad via Digitalmars-d wrote:
> On Tuesday, 1 December 2015 at 13:40:59 UTC, terchestor wrote:
> >botched this one. If Walter made this one everybody would threaten to leave D forever and want his head on a spike. But the mistake must be fixed anyway.
> 
> You seriously think DUB is D's biggest problem? You don't even need it to make use of D.

I find this whole debate hilarious, actually.  We aren't even talking about D itself, but about a tool that not everyone who uses D uses (I don't use dub and don't see any future scenario where I might need to use it), and that, a silly *configuration syntax* for said tool, and an *optional* one at that. Seriously, you can't get closer to Parkinson's law of triviality[1] than that. This isn't about semantics (that nobody understands nor cares about, being the most pertinent issue in a programming language and all that), not even the syntax of D itself, but about the (optional!) syntax of an optional(!) tool that some people use with D, and we're getting all worked up like the world is about to end or something.

Simply hilarious.

[1] https://en.wikipedia.org/wiki/Parkinson's_law_of_triviality


T

-- 
Why is it that all of the instruments seeking intelligent life in the universe are pointed away from Earth? -- Michael Beibl

On Tuesday, 1 December 2015 at 17:26:13 UTC, Andrei Alexandrescu wrote:
> On 12/01/2015 09:18 AM, Timon Gehr wrote:
>> On 12/01/2015 03:01 PM, Saurabh Das wrote:
>>> now told that 75% of the community doesn't like the change.
>>
>> I'm sorry to enter this discussion, but that's simply not what the poll
>> is saying. It should be very obvious that the poll is basically
>> meaningless.
>
> Independent on the topic at hand - wondering what your reasoning is. I just took a look and there are 205 votes. Not a large number, but quite a lot more than any voting we saw in the past (when consensus was proclaimed after like 15 votes :o)). Intuitively I agree with you, but I wonder at what point numbers become large enough to capture meaning. Thx! -- Andrei

One possibility is that you have to care enough to vote. It's hard to imagine that there is symmetry between those that dislike the new format and those that don't. And some members of the community don't even use Dub.

On other matters, 15 votes is fine, if they're the ones that care and have the background to form an opinion.

On Tuesday, 1 December 2015 at 17:26:13 UTC, Andrei Alexandrescu wrote:
> Independent on the topic at hand - wondering what your reasoning is. I just took a look and there are 205 votes. Not a large number, but quite a lot more than any voting we saw in the past (when consensus was proclaimed after like 15 votes :o)). Intuitively I agree with you, but I wonder at what point numbers become large enough to capture meaning. Thx! -- Andrei

Time for a proportion Z-test! Assuming that this poll represents a simple random sample (which it almost certainly doesn't, but it's the best we have), let's test the claim that at least 50% of the community dislikes the SDL format.

For reference, I'm using the poll data at the time of this post: 124 for, 145 against.

According to my calculations, at a 95% confidence level, there's not sufficient evidence to reject the hypothesis that at least 50% of the community dislikes the new format.

However, FWIW, we can say with very high confidence that at least 25% of the whole community likes the new format.