View mode: basic / threaded / horizontal-split · Log in · Help
September 12, 2004
Re: Most important issues for D (MIID)
"Joey Peters" <Joey_member@pathlink.com> wrote in message news:ci15ag$2rea$1@digitaldaemon.com...
> >Lots of people have commented that Phobos is crap. But I need specifics.
>
> What I think about it:
>
> No more c style things like time_t, printf, or that kind of things. It'd be also
> nice if it would be a whole fish rather than a school of fish.

D, like the other good non-commercially aligned powerful systems-software language, is not a one-shoe fits all. 
Thankfully, it never will be. If you want that, you're looking in the wrong place, and expend effort in going forward 
and petitioning for it will be a waste of your time.

> Also it looks
> like it's code could need a revamp. Containers should also be embedded, though I
> don't have a problem with that since I have a great time making my own
> containers ;)

What do you mean by embedded?

> D's standard functionality sort of patches that problem up, though
> that doesn't mean it doesn't need it. How common is it that you want to push
> items on an array? That you want to sort an array, or whatever?

It is not clear whether you are for or against.
September 12, 2004
Re: Most important issues for D (MIID)
"Sai" <Sai_member@pathlink.com> wrote in message news:ci0u0m$2k7g$1@digitaldaemon.com...
> Walter, I think I should apologise, I don't mean to be so harsh.
>
> Actually, I feel that the "D Language" is such an achievement,
> it is something which I have been looking for since last 7 years.
>
> Language is great, compiler is great, but somehow I feel that
> phobos got neglected.
>
> Specifically,
>
> 1. Where is DTL ?

news://news.digitalmars.com/digitalmars.D.dtl

> Why there are no direct links from website
> to DTL ?

Because it's not ready yet. In the next round I'll do a website as well, with Doxygenated help.

> Why should I dig NG for downloading DTL.

Because no-one's paying me to work on it full-time, more's the pity. Hence, there's been no (time to prepare a) website 
yet. If you browse the website, all the design intent is in there.

> Why can't
> we make DTL part of phobos ?

We will, eventually. At the moment we have two streams of development. DTL is working on "concepts" and MinTL is 
ploughing implementation furrows. This is actually a nice thing because we're learning lessons that don't overlap, and 
DTL may well be able to incorporate MinTL's containers (and their growing use/testing/robustness) into the DTL 
framework. I think that's a big plus!

You must remember that D's templates are an almost entirely new mechanism, and there are *no* experts in it anywhere. As 
such, no-one is in a position to write the definitive library that would go into Phobos at this time, and doing so would 
just piss people off as each release changed the interfaces to the library.

> No wonder so many versions of
> miniTL, xTl yTl's are floating around, infact I have my
> own version of TL in D !!
>
> 2. I feel that the amount of API is very less. The API which is
> currently in phobos seems to be there just to help you in
> writing compiler. it wasn't of much use to me. I had to
> write lot of classes when I tried to port my code to D.
> For example, we don't have 'Timer' classes yet.
>
> 3. Documentation is very bad. I think any powerful API is useless
> unless it is well documented. Lacking of "Search" facility
> or "Index" in D documentation is a huge discouraging factor.

True. Unfortunately we've not so far defined a documentation standard, though if you search the archives you'll see that 
many of us have been advocating Doxygen-compatible doc tags in all Phobos code (and all my submissions have it). Like 
many other things, time is the constraining factor. All I can say is that if we continue to mark up contributions, it 
can all be generated at a later time with relative ease.

> For D 1.0 above things are enough for me.
>
> Regards
> Sai
>
>
>
>
> In article <ci0gf5$18rn$1@digitaldaemon.com>, Walter says...
>>
>>
>>"Sai" <Sai_member@pathlink.com> wrote in message
>>news:chvfss$17j$1@digitaldaemon.com...
>>> 1. Put "D language and compiler" in maintainance mode ASAP.
>>> 2. Improve Phobos. Phobos right now looks like some
>>> bunch of crap put together.
>>
>>Lots of people have commented that Phobos is crap. But I need specifics.
>>
>>
>
>
September 12, 2004
Re: Most important issues for D (MIID)
>D, like the other good non-commercially aligned powerful systems-software language, is not a one-shoe fits all. 
>Thankfully, it never will be. If you want that, you're looking in the wrong place, and expend effort in going forward 
>and petitioning for it will be a waste of your time.

I thought D was designed with in mind that things together make sense
coherently, I'm not talking about making it fit for anyone. Should we end up
with bad conventions on everything, that even the runtime lib has inconsistancy?
I don't have a problem with this kind of behaviour if it wasn't shipped with D.
That whole statement is kind of vague anyway.

>> Also it looks
>> like it's code could need a revamp. Containers should also be embedded, though I
>> don't have a problem with that since I have a great time making my own
>> containers ;)
>
>What do you mean by embedded?

Into phobos/d release, doesn't really matter to me.

>> D's standard functionality sort of patches that problem up, though
>> that doesn't mean it doesn't need it. How common is it that you want to push
>> items on an array? That you want to sort an array, or whatever?
>
>It is not clear whether you are for or against.

It patches it up, in the sense of you can make it work with some effort. Though,
still, D seems to provide a whole lot of functionality that's not getting used
yet, also hence to people that just stick to their own ideas, programming around
those features. It's like Mars itself, if we could get there, we could do a
whole lot with it 'eventually' if we could convince ourselves it's not just a
gigantic red rock (why would we want to go there in the first place, because
it's a big red rock?). Phobos just only sees one part of mars at a time ;)
September 12, 2004
Re: Most important issues for D (MIID)
> >D, like the other good non-commercially aligned powerful systems-software language, is not a one-shoe fits all.
>>Thankfully, it never will be. If you want that, you're looking in the wrong place, and expend effort in going forward
>>and petitioning for it will be a waste of your time.
>
> I thought D was designed with in mind that things together make sense
> coherently, I'm not talking about making it fit for anyone. Should we end up
> with bad conventions on everything, that even the runtime lib has inconsistancy?

I haven't advocated that. Incoherence is not necessarily the obverse of multi-paradigm.

> I don't have a problem with this kind of behaviour if it wasn't shipped with D.
> That whole statement is kind of vague anyway.

I was trying to avoid impoliteness.

>
>>> Also it looks
>>> like it's code could need a revamp. Containers should also be embedded, though I
>>> don't have a problem with that since I have a great time making my own
>>> containers ;)
>>
>>What do you mean by embedded?
>
> Into phobos/d release, doesn't really matter to me.

Ok.

>>> D's standard functionality sort of patches that problem up, though
>>> that doesn't mean it doesn't need it. How common is it that you want to push
>>> items on an array? That you want to sort an array, or whatever?
>>
>>It is not clear whether you are for or against.
>
> It patches it up, in the sense of you can make it work with some effort. Though,
> still, D seems to provide a whole lot of functionality that's not getting used
> yet, also hence to people that just stick to their own ideas, programming around
> those features. It's like Mars itself, if we could get there, we could do a
> whole lot with it 'eventually' if we could convince ourselves it's not just a
> gigantic red rock (why would we want to go there in the first place, because
> it's a big red rock?). Phobos just only sees one part of mars at a time ;)

I'm afraid this is still entirely opaque to my terrestrial comprehension.
September 12, 2004
Re: Most important issues for D (MIID)
"Walter" <newshound@digitalmars.com> wrote in message news:ci0gf5$18rn$1@digitaldaemon.com...
>
> "Sai" <Sai_member@pathlink.com> wrote in message
> news:chvfss$17j$1@digitaldaemon.com...
>> 1. Put "D language and compiler" in maintainance mode ASAP.
>> 2. Improve Phobos. Phobos right now looks like some
>> bunch of crap put together.
>
> Lots of people have commented that Phobos is crap. But I need specifics.

There've been probably 75% of the specifics reported in the NG over the last, say, 12 months. However, I can't conjure 
them to mind now any more than you can. A series of directed reviews may be our best bet. I will start with Exceptions 
next week.
September 12, 2004
Re: Most important issues for D (MIID)
"Matthew" <admin.hat@stlsoft.dot.org> wrote in message news:ci1crf$2u2i$1@digitaldaemon.com...
>> >D, like the other good non-commercially aligned powerful systems-software language, is not a one-shoe fits all.
>>>Thankfully, it never will be. If you want that, you're looking in the wrong place, and expend effort in going forward
>>>and petitioning for it will be a waste of your time.
>>
>> I thought D was designed with in mind that things together make sense
>> coherently, I'm not talking about making it fit for anyone. Should we end up
>> with bad conventions on everything, that even the runtime lib has inconsistancy?
>
> I haven't advocated that. Incoherence is not necessarily the obverse of multi-paradigm.

I meant "single-paradigm". Obviously.
September 12, 2004
Re: Most important issues for D (MIID)
Sai wrote:

> Walter, I think I should apologise, I don't mean to be so harsh.
> 
> Actually, I feel that the "D Language" is such an achievement,
> it is something which I have been looking for since last 7 years.
> 
> Language is great, compiler is great, but somehow I feel that
> phobos got neglected.
> 
> Specifically,
> 
> 1. Where is DTL ? Why there are no direct links from website
> to DTL ? Why should I dig NG for downloading DTL. Why can't
> we make DTL part of phobos ? No wonder so many versions of
> miniTL, xTl yTl's are floating around, infact I have my
> own version of TL in D !!

cool - is your TL available? I'm curious what it's like, if you don't mind.

> 2. I feel that the amount of API is very less. The API which is
> currently in phobos seems to be there just to help you in
> writing compiler. it wasn't of much use to me. I had to
> write lot of classes when I tried to port my code to D.
> For example, we don't have 'Timer' classes yet.

Again, do you have a link for that?

> 3. Documentation is very bad. I think any powerful API is useless
> unless it is well documented. Lacking of "Search" facility
> or "Index" in D documentation is a huge discouraging factor.

Do you mean the documentation for phobos or all of D? For Phobos I just
search the web page using the Find command on my browser or I look at the
list of modules and pick the one that looks relevant. It can be out of
date, though. "See also" entries would be nice - so that for instance
std.file could say to see also std.stream and std.mmfile and vice versa.

> For D 1.0 above things are enough for me.
> 
> Regards
> Sai
> 
> 
> 
> 
> In article <ci0gf5$18rn$1@digitaldaemon.com>, Walter says...
>>
>>
>>"Sai" <Sai_member@pathlink.com> wrote in message
>>news:chvfss$17j$1@digitaldaemon.com...
>>> 1. Put "D language and compiler" in maintainance mode ASAP.
>>> 2. Improve Phobos. Phobos right now looks like some
>>> bunch of crap put together.
>>
>>Lots of people have commented that Phobos is crap. But I need specifics.
>>
>>
September 12, 2004
Re: Most important issues for D (MIID)
Sai wrote:
> Walter, I think I should apologise, I don't mean to be so harsh.

...

> 1. Where is DTL ? Why there are no direct links from website
> to DTL ? Why should I dig NG for downloading DTL. Why can't 
> we make DTL part of phobos ? No wonder so many versions of
> miniTL, xTl yTl's are floating around, infact I have my 
> own version of TL in D !!

Here's a summary:
http://www.prowiki.org/wiki4d/wiki.cgi?DTemplateLibrary

(It even has links.)

...

> 3. Documentation is very bad. I think any powerful API is useless
> unless it is well documented. Lacking of "Search" facility 
> or "Index" in D documentation is a huge discouraging factor.

Have you tried this:
http://www.digitalmars.com/advancedsearch.html

> 
> For D 1.0 above things are enough for me.
> 
> Regards
> Sai

-- 
Justin (a/k/a jcc7)
http://jcc_7.tripod.com/d/
September 12, 2004
Re: Most important issues for D (MIID)
>I'm afraid this is still entirely opaque to my terrestrial comprehension. 

Does the earthling understand the concept of never minding?

Nevermind :P
September 12, 2004
Re: Most important issues for D (MIID)
>It's time to prioritize and get us on a rational, sensible path for
>releasing 1.0.

Awesome!

-array arithmetic operations
-complete CV4 debug info

Thanks for including us!

Charlie


In article <cht9gl$20m8$1@digitaldaemon.com>, Walter says...
>
>In some delightful ways, D is a victim of its own success. There's a lot of
>traffic in the D newsgroups, sometimes getting to 100 messages per day. Not
>a day goes by without new proposals for features. It's simply beyond my
>capacity to give these all the attention they deserve, or even to read them
>all. I probably spend a minimum of 2 hours a day reading messages, and I
>could easilly spend 12 hours a day at it, and accomplish nothing else.
>
>Hence, I inadvertently overlook important issues. I know that this has been
>frustrating to some people, and I apologize for it.
>
>So I'd like to kick off this thread as an opportunity for all to post their
>two Most Important Issues for D with respect to getting 1.0 done. By MIID, I
>mean pragmatic things like:
>
>1) compiler bugs
>2) language shortcomings with no reasonable workarounds
>3) issues that are fundamentally blocking projects from using or proceeding
>with D
>4) severe library faults
>
>I don't mean things like:
>
>1) D 2.0 issues
>2) feature proposals like "It would be nice if ..."
>3) minor irritants
>4) philosophical issues
>5) issues that have been beaten to death already <g>
>
>If a thread here exists for the topic, a reference to that thread would be
>nice rather than reproducing it.
>
>It's time to prioritize and get us on a rational, sensible path for
>releasing 1.0.
>
>
1 2 3 4 5 6 7 8
Top | Discussion index | About this forum | D home