2010-04-27

google newsgroup, communication tech progress

2010-04-27

http://groups.google.com/group/comp.lang.lisp/browse_frm/thread/bc9f9552e3dae7e4

On Apr 27, 12:11 am, Juanjo wrote:
> On Apr 24, 8:08 pm, Francogrex wrote:
>
> > On Apr 23, 6:07 pm,XahLee wrote:
> > > then, click on the “Report as spam” for the 10 spam posts that shows
> > > on the first page.
>
> > It's useless; Google doesn't do anything significant to stop the spam.
> > This is evident from the fact that I and others have been reporting
> > those for a very long time and I even sent emails to google staff
>
> I have noticed the following: if I click on a "report spam" link, the
> RSS view of the group contains less spam. Today I came here and out of
> the 10 spam messages in the front page of Google Groups only three
> were in my RSS view in Google Reader. Coincidence? Probably not.

yes i agree that google care, and do something about this.

The click spam feature was added recently, perhaps 6 months ago, that indicates they do care, even if it is not that effective.

also, how effective it is depends of course on how many people report spam. For groups like comp.lang.lisp, many are old timers, don't use google, and the readership is already comparatively very small. So, it can get worse.

another thing might be interesting is that the spam rate seems to depends on the newsgroup too. For example, comp.lang.lisp is now 95% spam, but however, comp.lang.python gets almost no spam at least as shown in google group. This probably has to do with number of users, and also that comp.lang.python is mirrored with python's mailing list.

technology marches on by changing needs. In the 1990s or before, newsgroup is effectively the only medium and technology of subject oriented online forum. (besides a few commercial ones, e.g. CompuServe) Thru the years since 1990s which i personally lived thru, other tech of communication came into being that became widely adopted, roughly in chronological order: mailing lists, irc, faq-o-matic (a pre-cursor to wiki), Instant messaging, blogs, wiki, social network sites, youtube, twitter, and today much of these are all intermingled and inter-connected. For example, much sites that do any type of communication often has mailing list, web feed (rss/atom), web interface, instant messaging, all together as one integrated technology, not as much as independent technologies. (e.g. facebook, much of google's many services) Voice and video chat and conferencing is today almost everywhere too.

One point i would like to note that is, if google didn't provide the newsgroup service in 2001 (or the dejanews didn't start it), newsgroup might have gone the ways of dinosaur, much like many unix net tech such as who, talk, finger, gopher.

some wikipedia link for those curious
http://en.wikipedia.org/wiki/Dejanews
http://en.wikipedia.org/wiki/CompuServe

Xah
∑ http://xahlee.org/

2010-04-23

help save comp.lang.lisp from spam

today, comp.lang.lisp is 95% spam.

let's help save it. Regardless what newsgroup readers you use, when you do your daily newsgroup reading, please take 30 seconds to go to:

http://groups.google.com/group/comp.lang.lisp/topics

then, click on the “Report as spam” for the 10 spam posts that shows on the first page.

Thanks.

Xah
∑ http://xahlee.org/

do chicks need to show tits to indicate interest?

On Apr 22, 5:38 am, Adam Funk wrote:

> I'm fairly sure the factual claims in what I wrote are correct.  The
> opinion parts are where we disagree, ...

The number of recalcitrant donkeys online are amazing. Even after showing them proof, they insist to the contrary.

lol.

my problem has been how explicit should i be. You see, i refuse to be too explicit. Like, a chick needn't to pop her tits out in order to solicit intercourse. And if the guy still don't get it, the chick might as well change her mind.

so, at this point, my self-struggle is whether i should define, EXPLICITLY, a fact we are debating, then, cite references, and insert many examples EXPLICITLY in the post.

should i??

you see, even if we went to this trouble, agreement still may not come. Judging from past experience with tech geekers... the problem then becomes one of interpretation or philosophy. They say, o but that's not what i meant. Or, change of subject, like, “but there's no alternative”, or, “but you are just a troll”.

I LOL.

Xah
∑ http://xahlee.org/

top internet languages

On Apr 22, 4:23 pm, p...@informatimago.com (Pascal J. Bourguignon) wrote:
> Peter Axon writes:
> > [...] English [...] (the standard language of the Internet) [...]
>
> No, not really.
>
> http://www.internetworldstats.com/stats7.htm
>
> English is used by less than 1/3 of the Internet users, and less than
> 40% of the web pages.  And more importantly, its usage is dropping fast
> (relatively).

been keeping a eye on this for the past years, since about ~2005.

a more reliable data source on this would be
http://en.wikipedia.org/wiki/Global_Internet_usage

though, much data on the page is referenced to internetworldstats.com, which am not sure is very high quality.

am sure that Google must have published this on one of their blogs (e.g. web master blog)... but i cant remember any. Alexa.com might have that info too.

Here's related data:

• Unicode Popularity On Web
http://xahlee.org/comp/unicode_on_web.html

• Web Tech Stats 2010
http://xahlee.org/js/web_tech_2010.html

• Web Browser Market Share 2010-02
http://xahlee.org/js/browser_market_share.html

Xah
∑ http://xahlee.org/

2010-04-21

xah lee's writing valuation

2010-04-21

http://groups.google.com/group/comp.lang.lisp/browse_frm/thread/4e6a27ddc5ebf313#


On Apr 20, 8:56 pm, Peter Axon wrote:
> But you don't write in-depth technical stuff. You write in a
> blogger-tutorial style.

humm... that's a reasonable critique. Am trying to think hard, of all
my hyperbole and boasts of my writing skills and knowledge, i don't
think i have written a coherent academic quality work of more than,
say, 50k words. Well, actually, my special plane curves, my wall paper
group exposition, my emacs tutorial, come close, or very close. In
fact, thinking about these works, it simply thwarts your claim. They
are rather in-depth and not just blogging styled
tutorial. The plane curves and emacs each are book sized too, much
more than 50k words. Especially the math works, are cited in more than
ten academic journals and text books.

what am thinking more to myself, is that i don't think i have wrote
some none technical subjects that's substantial and coherently
sizable. All i have is pretty much a collection of a hundred or 2
disparate short essays that are more or less so-called creative
writings, mostly academically classified as rant style of social
commentary.

But here again, what am actually doing is close to introspection of
self-worth — a flaw of mine. All things considered, of my published
writings, with as much indifference as possible, i'd say they are more
worth than say lifetime output of more than 90% of today's PhDs.

> The following are from your article:
> > Do not use a "and" for the last item in a sequence of things, unless
> > it is too odd. For example, write "My favorite fruits are peach,
> > banana, cherry", not "My favorite fruits are peach, banana, and
> > cherry".
> > The article "an" is always written as "a".
>
> Which clown invents their own grammar? It makes your stuff just awkward
> to read.

Awkwardness is a point of view, young man.

Don't you find Shakespeare awkward to read?

Don't you find Chinese awkward to read?

Do you find one cuming or one Finnegans awkward to read?

have you thought about the reasons why? If you put your mind into this, you could, write a 3 thousands words essay about this topic. For example, here's some tips to start your essay. Classify the reasons. Give example of each. Quote the part of literature for each. Define what is meant by “awkward to read”. And, unavoidably, you probably have to philosophize about the audience. Depending on where you want to go, your essay can turn into classification of audiences, research report on english styles, nature and character of works from recognized stylists, definition of english styles, history of english styles, survey of world's styles by region... etc. (the work in this can become years long, tantamount to a phD's thesis.)

After which, you'll have a reasonable workout on this matter, then perhaps you'll see, my point of view.

alternatively, you can just read more of me.

• To An Or Not To An
http://xahlee.org/Periodic_dosage_dir/bangu/an.html

• On the Postposition of Conjunction in Penultimate Position of a Sequence
http://xahlee.org/Periodic_dosage_dir/t2/1_2_and_3.html

Xah
∑ http://xahlee.org/

2010-04-20

on writing, first draft vs perfection first

2010-04-20

m_mom...@yahoo.com (Mario S. Mommer) writes:
«In my experience, writing directly to TeX or anything similar is asking for trouble unless you are doing simple things. The source is not all that readable, and the hardcopy looks too good. The result is that you do not see the mistakes and the holes in the arguments. Taking a draft on paper and cleaning it up by copying the non-strike-out to new a paper draft is the only really good way to make sure you are really really really going over every detail again.»

I agree here.

«My senior year, I chose to write all my papers by hand, starting from an outline, then rewriting it fleshed out a bit, then rewriting again, and so on until I had a finished paper, all written longhand.  Then I'd go home and typeset it with LaTeX.»

lol. Writing by pen on paper? That's not a good advice. In fact, it would be too slow to be workable for me. For me, the ratio of ideas i need to put out vs the speed i can do with pen is not good, resulting in many thoughts lost in the hand laboring.

It is important to make writing a separate process from typesetting and or the technology used in writing. So, when i write math, or in general other subjects, the first draft i just quickly put out all my main ideas quickly, then, edit, format, typeset, and prettify.

For example, in the edit process, if the text is technical tutorial or exposition, i reduce difficult word, shorten sentences, make steps explicit, make statement precise, add code samples, more research for code snippets, references, etc.

If the writing is essay in literary context, i may for example add more fancy words or literary devices.

Actually, just yesterday i wrote:

• The Writing Style on XahLee.org
http://xahlee.org/Periodic_dosage_dir/bangu/xah_style.html

it describes the writing style i follow.

On Apr 19, 4:40 am, r...@rpw3.org (Rob Warnock) wrote:
Robert Uhl   wrote:

> As Halmos said in his classic little essay:
>
>    http://www.maths.ed.ac.uk/pg/data/halmosw.pdf [3.3 MB, 30 pages]
>     How to write mathematics
>     P. R. Halmos
>     ...
>     6. Write in Spirals
>     The best way to start writing, perhaps the only way, is to write
>     on the spiral plan. According to the spiral plan the chapters get
>     written in the order 1, 2, 1, 2, 3, 1, 2, 3, 4, etc. You think you
>     know how to write Chapter 1, but after you have done it and gone
>     on to Chapter 2, you'll realize that you could have done a better
>     job on Chapter 2 if you had done Chapter 1 differently. There is
>     no help for it but to go back, do Chapter 1 again differently, do a
>     better job on Chapter 2, and then dive into Chapter 3. And of course
>     you know what will happen: Chapter 3 will show up the weaknesses of
>     Chapters 1 and 2, and there is no help for it... etc., etc., etc.
>
> I especially have always liked this bit:
>
>     When you come to rewrite, however, and however often that may
>     be necessary, do not edit but rewrite. It is tempting to use a
>     red pencil to indicate insertions, deletions, and permutations,
>     but in my experience it leads to catastrophic blunders. Against
>     human impatience, and against the all too human partiality everyone
>     feels toward his own words, a red pencil is much too feeble a weapon.
>     [...] Rewrite means write again -- every word.

Can't say i support this idea at all. Such “spiral” method result in never finishing anything.

with this context, i'd recommend rather to actually complete a first draft of the whole, first. Then, come back and revise. The gist being that you want to get it done first, and never fuzz over on what you've already written.

knowledge and learning is a life long process. One always learns more and more. I'd say most authors would write a better book if they can do it again.

... which clown is it that gave this “spiral” shit? So i looked up:
http://en.wikipedia.org/wiki/P._R._Halmos

ok, a known mathematician. O well. Know that tremendously many specialized experts in history have given outrageously stupid advices in areas outside of their expertise. Not saying that his “spiral” method is such example... since all i know of it is the quoted paragraph here. Note also that Knuth has published a booklet (or so) about technical writing. (i think i scan'd it in early 1990s) Cant't say i was impressed in anyway by Knuth's writings... though it's long ago and i don't remember what it says, but usually i'd remember something about it if it made a impression on me.

Xah
∑ http://xahlee.org/

2010-04-16

emacs UI and doc problem: compute date in emacs

On Apr 16, 1:50 pm, José A. Romero L. wrote:
> On 16 Kwi, 20:23, Xah Lee wrote:> computing dates.
>
> > Recently i needed to compute dates in emacs.
>
> > For example, i need to find out:
> > 2010-04-16 - 215 days.
>
> (...)
>
> Something like this should do the job:
>
> (format-time-string
>  "%Y-%m-%d"
>  (time-subtract (date-to-time "2010-04-16 00:00") (days-to-time 215)))
>
> Notice that you *must* specify at least hour and minute (e.g. 00:00)
> in order
> for the date string to be parseable.
>
> Have a look at time-date.el for some useful time-related functions.

Thanks for both replies. After spending about 10 min on each suggested solution, still can't solve it, i end up google for a date calculator and and in 10 sec found
http://www.timeanddate.com/date/dateadd.html
and finished doing what i needed for 5 calculations of dates.

here's some complaints about the emacs solutions in case some emacs dev is interested.

The calender mode... so i type Alt+x calendar. Been using it for 10 years, but mostly only just to look at today's date and day of the week.

After Edward's suggestion, i thought yeah why didn't i thought of it,
cause i knew the emacs calender does all sort of esoteric calender
systems and thus must contain ways to do simple day substraction. But
while in calender, am not sure how to use it to computer yyyy-mm-dd
minus days. So i typed C-h m to read the online doc. It has close to
200 commands. I searched for string “diff” or “substr” but no
result. Then i looked up its info doc by typing ? while in the
mode. Quickly got to the node on Counting Days. Learned that i can
find the number of dates by selecting text in the calender. But other
than that, still not sure how to use it to solve my problem at
hand. At this point, i dropped this potential solution because i can
probably find other solutions faster than investing further time on
this one.

For Jose Romero's suggestion:

(format-time-string
"%Y-%m-%d"
(time-subtract (date-to-time "2010-04-16 00:00") (days-to-time 215)))

this code results in "1999-05-30". Apparantly a wrong result. Both
date-to-time and days-to-time function returns a weird result of a
list of 2 integers. e.g.

(date-to-time "2010-04-16 00:00") returns (14445 17280)

(days-to-time 215) return (283 29312)

looking up the inline doc of days-to-time, all it says is: “Convert
days into a time value.”. The inline doc of date-to-time is similarly
not helpful.

So, i looked up their package time-date.el, thinking that the header should provide some info on how to use this. Unfortunately, it doesn't.

The closest in this cryptic paragraph:

;; Time values come in three formats. The oldest format is a cons
;; cell of the form (HIGH . LOW). This format is obsolete, but still
;; supported. The two other formats are the lists (HIGH LOW) and
;; (HIGH LOW MICRO). The first two formats specify HIGH * 2^16 + LOW
;; seconds; the third format specifies HIGH * 2^16 + LOW + MICRO /
;; 1000000 seconds. We should have 0 <= MICRO < 1000000 and 0 <= LOW
;; < 2^16. If the time value represents a point in time, then HIGH is
;; nonnegative. If the time value is a time difference, then HIGH can
;; be negative as well. The macro `with-decoded-time-value' and the
;; function `encode-time-value' make it easier to deal with these
;; three formats. See `time-subtract' for an example of how to use
;; them.

according to this, i can see examples from the inline doc of the function
time-subtract.

But the inline doc of that just says:

Subtract two time values, t1 minus t2.
Return the difference in the format of a time value.

no info about what the format really means, what's the high low thing, or general outline of what are the major functions or explanation of the time format. How info about what this package is really for. Of course, it has to do with computing date and time, but how or why? what it adds to emacs default date time functions? how to use it? does it do the unix epoch seconds? Why is this package created? ...

At this point, i also stopped looking further with this potential solution.

Of course, if i persist, 30 min, 1 hour, 2 hours, or more, am sure will find out how to do the simple calculation i needed. But with cost considerations, both of these failed, even though i think they could easily be the best solution.

I think the above illustrate some emacs problems. Bad interface and or bad documentation. (i've been using emacs daily for about 12 years and coding elisp for 5 years)

... i've already done what i needed, but in retrospect, i can probably easily do it by converting my date into unix epoch seconds, minues the num of days i want to substract as expressed in seconds, then convert the unix epoch seconds back to yyyy-mm-dd. I can probably do this in elisp easily using builtin function, or i can do it with unix shell, or perl...

Xah
∑ http://xahlee.org/



------------------

http://groups.google.com/group/comp.emacs/browse_frm/thread/0daa90546fb4bfba/

On Apr 16, 9:42 pm, Jason Rumney wrote:
> In calendar mode, you can do it interactively.
>
> From the "Goto" menu, you can find "Other date" which has the keyboard
> shortcut "g d" (in your example case you don't need this step, since
> the date was yesterdays date). Then you can use C-u NUMBER_OF_DAYS
> to go back a number of days.

thanks a lot Jason. That does it. Pretty easy too.

> > For Jose Romero's suggestion:
>
> > (format-time-string
> >  "%Y-%m-%d"
> >  (time-subtract (date-to-time "2010-04-16 00:00") (days-to-time 215)))
>
> > this code results in "1999-05-30".
>
> In 23.1.1 (x86_64-pc-linux-gnu, GTK+ Version 2.18.3) of 2010-03-16 on
> crested, modified by Debian; the code above results in an error,
> because the seconds are required in the argument to date-to-time.

good info.

Xah
∑ http://xahlee.org/

2010-04-10

Paul Graham's Hacker's lang

2010-04-10

it's been brewing in my mind for a while to write a criticism on Paul Graham's Arc and his essay about ideal language.

of the various essays of his i read in the past years about Arc or designing of a lang, his essential idea lies on the concept of “hacker”. He keeps saying, lang needs to be this or that because “hacker” is this or that way.

That illusive thing, “hacker”. Whence, one can't really get a precise idea what he consider as ideal in a language design. I think that is the main problem, and consquently, whatever he comes up i can't deem good. (and from seeing what he actually have done with Arc, you know my opinion of it is shit, and as well it generally isn't well received... e.g. far less fanfare than say Haskell, Clojure, Scala, erlang, OCaml/F#... and far less users than say NewLisp, OCaml, Scheme Lisps...)

In 2008 i wrote a essay listing tens of new langs that are popularly talked about:

• Proliferation of Computing Languages
http://xahlee.org/UnixResource_dir/writ/new_langs.html

But in the past 2 years, more have come into the mass radar!!! Google's Go, Sun Micro's Fortress. All these langs have a bunch of philosophies and defense on how or why they should exist, with complaints about how they fill a hole. Some are dubious of course, but some do seem sensible. If you consider problem space of computing, and possibilities of imagination, it really has room for new langs.

Though, as far as i see, i can't see any concrete or theoretical merit of Arc with his “hackers need it!” concept.

Clojure for example, i can justify easily, for one thing, it is a lisp on JVM, modern lisp without baggage, easy install, independent, or with concurrency argument. NewLisp, i see merit, for example, it fills the scripting niche, and for hobbyist coders.

I really hate the word “hacker”. Imagine, what Dijkstra will have to say about “a lang designed for ‘hacker’?” LOL.

... what society overwhelmingly asks for is snake oil. Of course, the snake oil has the most impressive names —otherwise you would be selling nothing— like “Structured Analysis and Design”, “Software Engineering”, “Maturity Models”, “Management Information Systems”, “Integrated Project Support Environments” “Object Orientation” and “Business Process Re-engineering” (the latter three being known as IPSE, OO and BPR, respectively).”

— Edsger W Dijkstra (1930-2002), in EWD 1175: The strengths of the academic enterprise.

Xah
∑ http://xahlee.org/

combine TeX with a proof checking lang??

has On Apr 10, 5:53 am, m_mom...@yahoo.com (Mario S. Mommer) wrote:
> p...@informatimago.com (Pascal J. Bourguignon) writes:
> > What about if you merged LaTeX with an automatic proof checker?
>
> That would have to be one hell of a proof checker! It would have to
> understand natural language, and know what can be "safely assumed" as
> being implicit, given the audience. It has to be that way because a
> maths paper written like a formal spec of something is not going far,
> because of its intrinsic wetware incompatibility. The referees will
> absolutely hate you for it. And then you get the "way too technical"
> rejection. Which is a good thing, because if such a paper were accepted,
> nobody would read it.
>
> In short, that automatic proof checker has to be a really damn good AI,
> of the type Lisp was originally concieved to make happen.
>
> I mean, I'm all for such a program, but I'm a tiny bit skeptical :-)

it is impractical for TeX to be merged into a proof checking language. I mean, if it happens, the result would really have nothing to do with TeX.

However, the idea of combining a system that can display math formulas and proof checking, or with a so-called computer algebra system, and also as a computer language, is a common need, and has in fact been done to various degrees. Mathematica for example, is one. While, most computer algebra systems (such as Maple), or any math tools used in physical sciences, all have a math formula display system builtin to some degree (MatLab, Sage, MathCAD...)

for more detail of this, see:

• Math Notations, Computer Languages, and the “Form” in Formalism
http://xahlee.org/cmaci/notation/index.html

Xah
∑ http://xahlee.org/