November 23
On Monday, 23 November 2020 at 17:25:21 UTC, bachmeier wrote:
> On Monday, 23 November 2020 at 16:41:52 UTC, Ali Çehreli wrote:
>
>> Having said that, part of me is screaming "people who are interested in shiny stuff need not apply" but I am explaining to myself that it's the wrong attitude. :/
>
> I don't really agree with that sentiment, somewhat along the lines of Ola's argument.

Did you agree with me or not? :) I am Norwegian... could not parse... :]

I am not negative to javascript programmers, but one issue D has had in the past has been that people pull in different directions and that inevitably leads to a sense of "nothing I do can affect the outcome".

I am very optimistic about D right now because I now see that people are pulling in the same direction. DIP discussions are productive, in the sense that most people agree on the overall direction, if not the details. Walter impressed me greatly on DConf by underlining a "we feeling" in everything he said, community management is very tricky and words and phrases matters. Whether @live will pan out or not is not important, what is important is that it is on the table, that means somebody else (maybe a master student) will think that they can produce a solution that is more complete and expect it to be accepted.

A massive amount of people with a different mindset (C#) could jeopardise the current positive trend of pulling in the same direction. So my personal viewpoint is that D's future does not depend on the number of users, but the number of enthusiasts that can pull in the same direction and make dents in the fabric with their own hands. So more people that share goals is positive, but more users, for the sake of numbers, could be destructive.

The compiler codebase has also improved quite a lot since it was in C++, so it is easier to experiment it. That should be encouraged. In the past people could be flamed for experimenting with the compiler in exotic ways, thankfully that is a thing of the past. And that is also a thing that really needs to be catered for (tutorials/documentation).

November 23
On Monday, 23 November 2020 at 09:53:29 UTC, ddcovery wrote:
> [...]

I don't see how Go's website looks anymore modern than D's. Further, Rust's landing page looks awful: the colors and layout make me nauseous. Typescript's just looks more angular. Etc. I'm not saying D's web presentation is perfect, but neither do I see any glaring flaws compared to the "competition." (If somebody wants to take the initiative and do a design overhaul, however, more power to them.)
November 23
On Monday, 23 November 2020 at 18:10:58 UTC, Ola Fosheim Grøstad wrote:
> On Monday, 23 November 2020 at 17:25:21 UTC, bachmeier wrote:
>> On Monday, 23 November 2020 at 16:41:52 UTC, Ali Çehreli wrote:
>>
>>> Having said that, part of me is screaming "people who are interested in shiny stuff need not apply" but I am explaining to myself that it's the wrong attitude. :/
>>
>> I don't really agree with that sentiment, somewhat along the lines of Ola's argument.
>
> Did you agree with me or not? :) I am Norwegian... could not parse... :]

I don't think the problem is that you're Norwegian. I don't write carefully in a setting like this. I can barely parse what I wrote. I was agreeing with you, for the record.


> I am very optimistic about D right now because I now see that people are pulling in the same direction. DIP discussions are productive, in the sense that most people agree on the overall direction, if not the details.

It took a few years (or decades) to figure out how to run the project. Now it's on track.
November 23
On Mon, Nov 23, 2020 at 07:47:31PM +0000, starcanopy via Digitalmars-d wrote:
> On Monday, 23 November 2020 at 09:53:29 UTC, ddcovery wrote:
> > [...]
> 
> I don't see how Go's website looks anymore modern than D's. Further, Rust's landing page looks awful: the colors and layout make me nauseous. Typescript's just looks more angular. Etc. I'm not saying D's web presentation is perfect, but neither do I see any glaring flaws compared to the "competition." (If somebody wants to take the initiative and do a design overhaul, however, more power to them.)

TBH, I find Rust's website design awful.  It follows that horrible modern trend of giving no helpful information on the landing page, and instead filling 90% of the screen with the modern equivalent of giant in-your-face banner ad, complete with oversized fonts and excessive emphasis.  Just about every page features this horrible information minimization, and it takes an excessive number of links before you get to the *real* information.

Typescript's website is slightly better, even though it suffers from the same in-your-face design philosophy, but at least the font sizes are not as excessive as Rust's.  It still takes an excessive number of clicks to get to the meat, though.

D's website isn't all that much better, unfortunately.  That fast-fast-fast slogan honestly makes me cringe and want to throw up. It also features "Support the D language" right in the bottom half of the page: if I'm a newcomer, why would I even want to support D when you haven't even told me why I should care.

TBH, all 3 websites suffer from the ridiculous modern trend of an essentially contentless splash page that tries to sell me something I haven't even understood yet.  The slogan-based design, e.g., Rust's "Why rust? Performance, reliability, productivity", tell me *nothing* about Rust beyond marketing buzzwords that I don't even read, because how do I even know how reliable such statements are?  It reads just like a marketing brochure that people hand me at a fair and that I promptly chuck into the trash can without even reading.

Show me *actual Rust code*, and describe to me *specific Rust features*, then I might be a bit more sympathetic.  Give me a list of features that I can compare against other languages in order to evaluate whether Rust offers me something other languages don't.

Overcoming my initial nausea, I click on "Get started", and I'm taken to a page where again, 30% of the page is occupied by the title of the page. What a useless waste of space!  Not to mention that horrible red-and-purple color scheme.  The nausea returns.  There's not a single line of Rust code anywhere that I can see. Just some obscure curl command without an obvious description of just what it does.  Why would I even want to install this when you haven't even shown me a single line of Rust?  I'm not sold yet, and you're already asking for my wallet.

TypeScript's website doesn't fare much better.  But it does have a major plus over Rust's website: it shows you *actual code*, along with an error message that shows exactly what TypeScript's selling point is. And there's a "try in your browser" link that straightaway lets me try it out for myself (which would've been great, but my browser has JS disabled for new sites by default, so that resulting empty page was not exactly a positive impression).  But again, 30% of the front page is filled with marketing spiel that I don't even bother to read.

D's website at least doesn't have nausea-inducingly huge fonts on the front page, and there's also actual code to boot. A win over Rust's website.  I can edit and run code *right on the front page*: another plus.  Unfortunately, that "support the D language" section is just out of place.  Support usually comes from people who have *already* been sold on D, and such people usually already know the ins and outs of the website, where to look for information on how to support the language, etc.. Such information *does not need to be on the front page*.  It just looks out-of-place to a newcomer.

D's website still suffers from the influence of that horrible splash page design philosophy. That paragraph about "general-purpose programming language with blah blah blah didn't bother to read" doesn't accomplish very much IMNSHO.  And too much white space on the page.  I can understand having adequate space so it doesn't look like a cluttered newspaper ad page, but still. This amount of whitespace on a page, so prevalent these days, is just ridiculous.

I'd much rather see, e.g., a list of features that make D stand out from other languages, right on the front page. This goes for any language, in fact. Then I can glance at it and know in 3-5 seconds what are the major features of this language, which are hopefully also what differentiates it from the crowd of other languages, without having to click on this or that or the other (which involves mental effort to process: which of those non-informatively named links will actually tell me what this thing is about in the first place?! -- your average net surfers are lazy, don't forget that).

Also, that "industry proven" section seems poorly placed: neither prominently at the top or deal-sealing at the bottom, but an indecisive middle.  I mean, why would I care who uses this language, if it doesn't have the features I want?  And following this, the remainder of the page just seems to be ad hoc, and rather disorganized. The order of the sections don't seem to follow any obvious logical order.  First there's a smattering of random links to news, community, forums, then that horrible fast-fast-fast slogan with more elaborate explanations at the very bottom (huh? shouldn't this be nearer to the top?).  Very strange.

The way these websites are designed, esp. Rust's, give me the impression that they're designed for CEO's and other management types who want to be wow'ed by how beautiful and in-your-face the website is, and based on that decide that their employees should use this language from now on, rather than the actual programmers who will be actually using the language, who:

1) Already know they're looking at a *programming language*, not brand name shampoo, and therefore are not scared of seeing actual code;

2) Already know at least one other programming language (we're not trying to teach non-programmers how to program, after all), and therefore have no reason to look further unless they see a feature that their current language doesn't have, or a feature that might be superior to their current language's features;

3) Are probably looking at your website because they're looking for a better language than they currently have, and therefore want to know right away what makes your language stand out from the hundreds of other languages out there;

4) May be existing users of your language who are looking for documentation or the latest release, or maybe they have a question they want to ask somebody (in the forum, etc.), so the faster they can get to this information, the better -- preferably 1 click away, and preferably *without* having to scroll.  I think everyone here already knows that very few people actually bother to read anything past what's shown in the browser window upon landing on the page.


The big question behind all of this is: who's the target audience of the website?  The people who will be using the language, people who are already using the language, or management types who want to be wowed with a flashy page that captivates their attention, or some generic, non-descript, abstract audience that may or may not exist?

Hopefully not the last two.  And of the first two, IMO the first is more important, because the second group is already sold on your product, so they're naturally more inclined to put in more effort to find the information they want.  And the primary question with the first is: (1) what's this language about (what are the defining features), and (2) why should I care (what features does it have that my current language doesn't have / what makes it better than what I have now).

//

Disclaimer: What I look for in a website is likely very different from what your average web surfer looks for.  I look for *information*, and presentation is secondary; whereas most people are wowed by presentation, then realize there might actually be information.  So YMMV.  But also: if you only cater to the whims of a fickle average web surfer, don't be surprised if they don't stay.  The kind of audience you sell to is the kind of users you'll get.


T

-- 
Why waste time reinventing the wheel, when you could be reinventing the engine? -- Damian Conway
November 23
On Monday, 23 November 2020 at 09:53:29 UTC, ddcovery wrote:
>
> ...

I think the home site is fine. It simple, fast and responsive. Documentation itself can be out of date though. The biggest problem I have is *this* what you are using right now, the newsgroup interface. It is so old and user unfriendly. There several forum SW to choose from out there so there is no reason to migrate. I just hate it when I see some mistake I've made but I cannot edit my post because this is the 90s.
November 24
On Monday, 23 November 2020 at 23:27:22 UTC, IGotD- wrote:
> On Monday, 23 November 2020 at 09:53:29 UTC, ddcovery wrote:
>>
>> ...
>
> I think the home site is fine. It simple, fast and responsive. Documentation itself can be out of date though. The biggest problem I have is *this* what you are using right now, the newsgroup interface. It is so old and user unfriendly. There several forum SW to choose from out there so there is no reason to migrate. I just hate it when I see some mistake I've made but I cannot edit my post because this is the 90s.

dlang.org website is really fine for D developers: it acts as a HUB where presentation, information, resources, forums, ... are integrated.   This is really a hard (and great) job.

And it's fast!!!

I have to clarify that I use, primary, linux (ubuntu) and this is the origin of the problem:  Dlang offers 3 download buttons (2 red, 1 white) with "complex" texts,  and this "breaks" completly the balance of the intro section (Windows users see only 1 "Downloads" button and intro section is well balanced).


With small changes you can "balance" the experience.

i.e.: (for linux)

* Use only 1 download button with a simple text "Download D"... this navigates to the download page where you offer your platform buttons (i.e,:  DEB an RPM) and the complete alternatives list.

If you are using Ubuntu, try this (on browser console) in dlang.org page:

$(".pitch").css('fontWeight',"350");
$(".download a.btn.action:first-child").css({"fontSize":"16px", "padding": "0.7em"}).html("Download <b>D</b>").attr({"href":"download.html"});
$(".download a.btn.action:not(:first-child)").remove();
$(".download a.btn:not(.action)").remove();

If your are using "Windows"... web site looks fine... May be an small change :-)

$(".download a.btn:first-child").
css({
  "fontSize":"16px",
  "padding": "0.7em",
  "border-color": "#98312A",
  "background-color": "#B03931",
  "color": "white"
}).
html("Download <b>D</b>").
attr({"href":"download.html"});



November 24
On Monday, 23 November 2020 at 09:53:29 UTC, ddcovery wrote:
> I noticed that kotlin, scala, typescript, rust, ... all of them take care about the "first impression" offering a modern web interface in the main site.  DLang website first impresion is "old", mainly because de size/location/color of the 3 download buttons, the height of the top navbar, and

Apologize for the SPA fad.
November 24
On Tuesday, 24 November 2020 at 11:30:55 UTC, Kagamin wrote:
> On Monday, 23 November 2020 at 09:53:29 UTC, ddcovery wrote:
>> I noticed that kotlin, scala, typescript, rust, ... all of them take care about the "first impression" offering a modern web interface in the main site.  DLang website first impresion is "old", mainly because de size/location/color of the 3 download buttons, the height of the top navbar, and
>
> Apologize for the SPA fad.

Well, I really like Rust website (I have not to apologize) and Go last months "light" changes are good.  Kotlin really want to be seen as "dynamic, young, smart",   Scala play framework website is nice... postgres decided to change their website too:  each one decides how to convince newcomers and how to compete in development market.   I think it has been a good discussion thread where the "D programmers like their website" is the main conclusion.  May be, the bad part of some opinions is "...and they have nothing to learn about other great developmet SPA fad sites".

This was my opinion:
>> DLang website first impression is "old", mainly because de size/location/color of the 3 download buttons, the height of the top navbar, and the difficult to see the missing title "The D programming language" or "D happy developers language" ( :-p )

Kagamin, I have not to apology for my respectful opinion about look & feel (not about hard and good website work) and I really appreciate to read and know other developers point of view.

November 24
On Monday, 23 November 2020 at 23:27:22 UTC, IGotD- wrote:
> I think the home site is fine. It simple, fast and responsive. Documentation itself can be out of date though. The biggest problem I have is *this* what you are using right now, the newsgroup interface. It is so old and user unfriendly. There several forum SW to choose from out there so there is no reason to migrate. I just hate it when I see some mistake I've made but I cannot edit my post because this is the 90s.

Sorry, but *I* am personally very happy that you can't edit.
I hate if someone says some bullshit an later on claims he never has said that. And if you try to prove, the culprit is actually edited away.

If you made a mistake, simply write a second post and correct it. If that's so '90s, I would be glad to live in the '90s again. Go home with your fake news, that's so 1984.
November 24
On Monday, 23 November 2020 at 23:22:14 UTC, H. S. Teoh wrote:
> [...]

I agree with most if not all of your points. I especially do not like the "industry proven" section as I think it overstates D's presence or is outdated (e.g. sociomantic and Remedy).
As you mentioned, restricting the landing page to require minimal scrolling will, hopefully, force one to consider what's really necessary to know about D. Also slapping a prominent link to D's tour (It's merely mixed in the middle as it is now.) somewhere on the page would do wonders. After all, what better way to signal a productive language to somebody than with readily available access to code?
Next ›   Last »
1 2