Re: standard Haskell

[John Hughes]

From [EMAIL PROTECTED]  Thu Dec 11 17:23:40 1997
Return-Path: <[EMAIL PROTECTED]>
Received: from mulga.cs.mu.OZ.AU (mulga.cs.mu.OZ.AU [128.250.1.22])
by animal.cs.chalmers.se (8.8.5/8.8.5) with SMTP id RAA10360
for <[EMAIL PROTECTED]>; Thu, 11 Dec 1997 17:23:37 +0100 (MET)
Received: from mundook.cs.mu.OZ.AU by mulga.cs.mu.OZ.AU with SMTP 
(5.83--+1.3.1+0.51)
id AA04744; Fri, 12 Dec 1997 03:23:07 +1100 (from 
[EMAIL PROTECTED])
Received: (from fjh@localhost) by mundook.cs.mu.OZ.AU (8.8.5/8.7.3) id 
DAA18741; Fri, 12 Dec 1997 03:23:06 +1100 (EST)
Message-Id: <[EMAIL PROTECTED]>
Date: Fri, 12 Dec 1997 03:23:05 +1100
From: Fergus Henderson <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Subject: Re: standard Haskell
References: <[EMAIL PROTECTED]>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 0.88
In-Reply-To: <[EMAIL PROTECTED]>; from 
[EMAIL PROTECTED] on Sun, Aug 24, 1997 at 12:39:21PM +0200

On 24-Aug-1997, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
> 
> You can see the ongoing discussion on
> 
>   http://www.cs.chalmers.se/~rjmh/Haskell/Display.cgi?id=0

Actually it is at

http://www.cs.chalmers.se/~rjmh/Haskell/Messages/Display.cgi?id=0

But it is difficult to track the ongoing discussion, because
- the interface is slowww (they don't call it the "World Wide Wait"
  for nothing)
- it is difficult to keep track of which parts you have read already
  and which parts are new
- unlike say a mailing list, those wishing to track the discussion must
  remember to check the Web site regularly (or to use the jargon,
  it's "pull" technology rather than "push" technology).

I spend some time looking at the Web site tonight, but eventually I got
sick of the net lag and gave up.

My question is this: was it the intent of the committee to deliberately
discourage the participation of those not on the committee?
Or was this feat achieved by accident?

-- 
Fergus Henderson <[EMAIL PROTECTED]>   |  "I have always known that the pursuit
WWW:    |  of excellence is a lethal habit"
PGP: finger [EMAIL PROTECTED] | -- the last words of T. S. Garp.

Re: standard Haskell

[John Hughes]

Fergus Henderson says:

But it is difficult to track the ongoing discussion, because
- the interface is slowww (they don't call it the "World Wide Wait"
  for nothing)
- it is difficult to keep track of which parts you have read already
  and which parts are new
- unlike say a mailing list, those wishing to track the discussion must
  remember to check the Web site regularly (or to use the jargon,
  it's "pull" technology rather than "push" technology).

I spend some time looking at the Web site tonight, but eventually I got
sick of the net lag and gave up.

My question is this: was it the intent of the committee to deliberately
discourage the participation of those not on the committee?
Or was this feat achieved by accident?

Certainly not! The is the first time the Haskell committee has ever conducted
its discussions in public, and the purpose is of course to encourage
contributions from others. The committee uses just the same interface as
everyone else, by the way, so you can at least take comfort in the fact that
everyone suffers alike! The only difference is that messages can only be
*posted* by a committee member. But others are welcome to contribute via
any member of the committee. A number of people have done just that. I think
it's essential to restrict additions in this way: would you entrust the
responsibility for a language design to an unmoderated newsgroup?

The interface is designed to present each message in the context of the
preceding discussion, and thus to help people make thoughtful contributions
based on the entire discussion, rather than just react to the last message. In
that sense it's geared towards the committee, who have a responsibility to
follow the entire discussion, rather than to the occasional visitor.

It's true that the first page has become rather long, and can take a few
seconds to download. Because it's produced by a CGI script it can't be cached,
which may be more of a problem if you're sitting in Australia than it is here
in Sweden.

The software that manages the Standard Haskell pages isn't necessarily
fixed. It's a collection of small Haskell programs which I have constructed
myself. I do occasionally enhance it in response to suggestions from the
committee -- for example, the addition of colour to identify recent
additions. I can enhance it to make it more useable for other people too
(although the time I have available for that is very limited).

Here are three things I could probably do if there's strong demand for them:

(1) Make the first page cacheable.

(2) Provide an alternative interface that displays each message with its
*immediate* children in the tree, but not their descendants. Harder to
navigate in, but at least the first page would be much shorter and quicker
to download.

(3) Provide a way to register your email address with the system. New messages
would be mailed to everyone in the register. (The committee already gets
this service, so it would be quite easy to add for others). I wouldn't
sell your email addresses to junk advertisers, honest!

But maybe, Fergus, you just need to buy a faster modem (:-)

Re: standard Haskell

[Ron Wichers Schreur]

Fergus Henderson wrote (to the Haskell Mailing List):

> [..]
> But it is difficult to track the ongoing discussion, because
> - the interface is slowww (they don't call it the "World Wide Wait"
>   for nothing)

I tried it yesterday and had no complaints about the performance.

> - it is difficult to keep track of which parts you have read already
>   and which parts are new

The colour of the author's name changes according to the age of the
message (from red to blue). Perhaps you don't have a colour monitor?


Cheers,

Ronny Wichers Schreur

Re: standard Haskell

[Jon . Fairbairn]

On 11 Dec, Paul Hudak wrote:
> I suppose that one improvement that you'd like and that I agree would be
> an improvement is the ability to mark messages as read. 

With Netscape Navigator (at least on Linux) you can set an option not
to expire visited links.  This means they change colour and stay that
way indefinitely.  I think 'indefinitely' here means 'until something
goes wrong with nerdscaphe'.

I suppose John could implement something using cookies, but why should
he put in so much effort?

  Jon

-- 
Jon Fairbairn [EMAIL PROTECTED]

Re: standard Haskell

[Paul Hudak]

Having participated in many previous Haskell design efforts, I must say
that John's WWW-based system is MUCH better than straight email.  With
email you have 16 different threads that are really hard to keep track
of; the tree-based approach keeps things better organized.  A newsgroup
isn't as good either, even if threaded.

Another point is that my strategy for interaction is different; it's not
daily, as it tends to be with email.  Rather, I pick an hour or two per
week (or whatever) to devote to the discussion; as a result I think I'm
more efficient.  Also, John's latest addition of "colors" should help
identify the most active dicussions.

I can believe, however, that if you are just joining the discussion you
may find it a bit overwhelming.  But what if you joined an email
discussion and were given an archive of umpteen messages?

By the way, I don't have any problem with WWW delays, even from home
over my 28.8K modem.  Fortunately the link is to Sweden, not to the UK,
which is notoriously slow from the states.

I suppose that one improvement that you'd like and that I agree would be
an improvement is the ability to mark messages as read.  I use the fact
that my browser changes the color of the entry to help me with this, but
that is lost next time around.

In any case, the committee certainly did not "deliberately discourage
the participation of those not on the committee"; indeed I'd say the
opposite strategy was taken.  I think it's very unusual for a committee
to open its dialogue to the world.

  -Paul

Standard Haskell Opinion Poll

[John Hughes]

Many thanks to everyone who has registered their opinion on
strict vs lazy pattern matching in the poll I'm conducting.
If you plan to do so but haven't yet, there's still time:
I shall freeze the results tomorrow.

For more information see

http://www.cs.chalmers.se/~rjmh/Haskell/Messages/Polls/matching.html

John Hughes

Lectureship at Umeå university

[Lennart Edblom]

Applications from readers of the Haskell list are encouraged!

Umea University announces a Lectureship in Computing Science.

Umea University in northern Sweden is a young and dynamic university with
more than 22,000 students and 3,900 employees. The Department of Computing
Science has experienced substantial growth in recent years, and is now one
of the larger departments of the university, with nearly 400 students
majoring in computing science. Members of the department are engaged in
research in a wide variety of subdisciplines of computing science, both
theoretical and applied. We are also engaged in undergraduate education in
several programmes of study.

We now welcome applications for a position as Senior Lecturer in Computing
Science. A main responsibilty for this lectureship will be teaching at the
undergraduate and master's levels. We have a need for teachers in "core
computer science" areas, specifically programming methodology, computer
networks and distributed systems, operating systems and computer
architecure, and database systems.

Teaching responsibilities will amount to approximately four courses each
year. The position also includes opportunities for research, as well as for
the direction of master's and doctoral students.

For further information, please contact Professor Bo Kagstrom, Chairman of
the Department of Computing Science, phone: +46-90-786 5419, e-mail:
[EMAIL PROTECTED], or Professor Patrik Eklund, phone: +46-90-786 9914, e-mail:
[EMAIL PROTECTED], or Lennart Edblom (director of studies), phone
+46-90-786 6137, e-mail [EMAIL PROTECTED]  For the complete announcement of
the position, see http://www.cs.umu.se/open_pos.

Applications should include a curriculum vitae, copies of degree
certificates, a statement of previous research achievements and teaching
merits, a list of publications, and reprints numbered according to the
list, all in two copies. The application should be marked dnr 3152-2050-97
and sent to Registrator, Umea University, S-901 87 Umea, Sweden. Closing
date is January 7, 1997.
---

Lennart Edblome-mail: [EMAIL PROTECTED]
Department of Computing Science...Tel: +46 90 786 6137
University of Umea... Fax: +46 90 786 6126
S-901 87 Umea, Sweden