Darkstar. Don't remember Spence. Ran EBBS-64, modded for 80 col display, quoted replies, rudimentary threading, and FTSC-001 packets. I remember stacks of SFD-1001s clicking all night long. Still have nightmares about the 8250 I smoked.
These people behave irrationally not because of the game, but because they are irrational, sick, or sociopathic people.
If these same individuals were in a knitting club, they'd be stabbing each other's eyes out with knitting needles and paying stupid amounts of money for fancy-assed wool to turn into butt-ugly sweaters and scarves. But we don't hear people telling us that knitting is evil -- probably because other people outside the knitting community understand what it's all about.
> Then you have a brokeass CNAME entry which goes against the RFC's, if I recall.
You don't really know much about DNS, do you?
www.edisonnews.com is a CNAME for d083489.edisonnews.com. There is nothing "broke ass" about it.
edisonnews.com has no DNS entry whatsoever (excluding SOA, obviously).
> A browser will get to the right place without that dumbass > "www" tacked to the front of a domain name.
Who says that machine d083489 is the right place?
Additionally, even if edisonnews.com. CNAME d083489 was inserted in the DNS, the site *still* might not work -- it would only work if www.edisonnews.com was the default site for that web server. Otherwise, he'd had to add another vhost to his httpd.conf-equiv. And given that he's running IIS, that might be a chore.;)
Incidentally, what RFC w.r.t CNAMEs are you talking about? The only relevant guidelines which pop to my head are an RFC in the 1300-range or so (been a while) discussing the use of CNAMEs with MX records and such. (MX to CNAME ist verboten, but has nothing to do with web traffic)
IIRC, the Wavecom WISMO has been doing this for years. I have several devices based on this chip, all they need is a different user interface to be fully-functioning phones.
> Google doesn't have magic hardware, they aren't reading through > 800MB of your e-mail every time you do a search.
Right. They don't have magic hardware, they have (effectively) magic software. And tonnes of hardware.
It may be possible to find a good mail indexing program on the client-side -- but then I'd have to download an index huge amounts of mail every time switched workstations. Not to mention that I'd probably be stuck on a Win32 solution, since ~80% of the workstations I use are Win32... I like being able to use Macs or the ultra 5 on my desk at work for e-mail every now and then. Especially when I want to cut and paste from an active xterm into an email..
I have yet to see an IMAP server with searching which even vaguely approaches GMail's speed, although I must admit, I haven't looked in a couple of years.
Server-side side searching is a fine option, provided you have a hell of a fast server when searching *huge* amounts of mail. Which I do regularly.
I don't know how GMail does it, but I can search through ~700 MB of email in ~2s (one one thousand, two one thousand, BING!).
It would require a LOT of effort to put together a server able to do that. Hell, it would require a LOT of effort to put together a server able to read from disk that quickly.
Not that I think AJAX is the magical solution, but it's a fine solution for that app -- heavy server, light front end -- and Google has taken away the "hard part" of figuring out how to get fast searches. It used to take me in excess of 5 minutes to do what they can do in 2s. For that convenience, I will "pay the price" of having the "wrong" From: address (but I'd pay to fix that if I could...)
But you're right, it's not faster because the network pipe is faster than my disk (duh!) -- it's faster because the back end is SO MUCH faster at searching and sorting than ANYTHING I've used on my desktop.
My typical mail usage pattern is to search through ~800MB of mail, three or four times a day, from several different boxes. The little delays when clicking "compose" and "reply" are really non-issues, compared to the time savings (over a minute per search).
What I like *BEST* about GMail is that it works (for me) as well as a desktop client (except from the From: header, GRRR!),but I can use it from *anywhere* -- and it's a DAMNED sight faster than IMAP!
Add IMG { behavior: url(/path/to/pngbehavior.htc); } to your CSS.
IE now supports transparent PNGs (but only for image tags). Remember to include height/width attributes when you're done developing, or it'll be a little flaky ('specially when image is already cached).
If it makes you feel any better, I write fairly DOM/CSS/Javascript heavy web apps, and test on IE5.5 and Firefox 1.0.
It's not hard, and I don't do *any* UA detection*. The key is picking the right standards subsections, and implementing the missing stuff yourself. IE has been "good enough" since version 5.5, assuming your layout won't get broken due to text in a DIV being "off" by a few pixels, etc. Remember, users come for content, anyhow!
Incidentally, most of my stuff works on IE4.0, but it's pretty damned ugly.
* UA detection is used to pop up an alert box bitching about the browser version and recommending firefox. Does not otherwise impact site. Non-JS users get bitched at via NOSCRIPT tags.
I don't know jack about developing under Windows, but cURL is a damn fine access library for the Unix folks. Perhaps you should look into it, it probably runs under windows (bsdsockets and wsock32 are similar enough to make porting trivial if it doesn't).
Not to put too fine a point on it, but I've been serving up gzipped content for a REALLY long time... let's see... I remember that the app I originally deployed it for was running under NS4/IE4, and those were the latest browsers at the time...
IIRC, NS and IE used different Content-Encoding mime types. I think one was "x-gzip" and the other was "gzip". But really, this has been supported *forever*. I don't understand why something like mod_gzip isn't built right into Apache (1.3 -- haven't changed to 2.x yet, so if it's there, I wouldn't know).
Slashdot Mirror
http://maps.google.com/maps?q=new+orleans+transit& ll=29.967985,-90.088888&spn=0.002199,0.003720&t=e& num=10&start=0&hl=en
.. oh wait, nevermind.
Good thing nobody tried to use them for the evacuations, they'd be
I wonder where all the school busses are parked?
Darkstar. Don't remember Spence. Ran EBBS-64, modded for 80 col display, quoted replies, rudimentary threading, and FTSC-001 packets. I remember stacks of SFD-1001s clicking all night long. Still have nightmares about the 8250 I smoked.
This is nothing more than more of the same crap that surrounded Dungeons & Dragons in the late 70s and the 80s.
r iticism_and_controversies
http://en.wikipedia.org/wiki/Dungeons_&_Dragons#C
These people behave irrationally not because of the game, but because they are irrational, sick, or sociopathic people.
If these same individuals were in a knitting club, they'd be stabbing each other's eyes out with knitting needles and paying stupid amounts of money for fancy-assed wool to turn into butt-ugly sweaters and scarves. But we don't hear people telling us that knitting is evil -- probably because other people outside the knitting community understand what it's all about.
WHAT?
;)
Shit.
Thanks. Glad I keep backups on another server!
No, wait, I got mail two hours ago, and that's all that I'm expecting (checked against backup).
Does this maybe only happen when you didn't sign up with the dots in your alias?
> I certainly hope this is a temporal problem and emails are not lost.
So, you think Google will time-travel to retrieve your lost e-mails?
> Then you have a brokeass CNAME entry which goes against the RFC's, if I recall.
;)
You don't really know much about DNS, do you?
www.edisonnews.com is a CNAME for d083489.edisonnews.com. There is nothing "broke ass" about it.
edisonnews.com has no DNS entry whatsoever (excluding SOA, obviously).
> A browser will get to the right place without that dumbass
> "www" tacked to the front of a domain name.
Who says that machine d083489 is the right place?
Additionally, even if edisonnews.com. CNAME d083489 was inserted in the DNS, the site *still* might not work -- it would only work if www.edisonnews.com was the default site for that web server. Otherwise, he'd had to add another vhost to his httpd.conf-equiv. And given that he's running IIS, that might be a chore.
Incidentally, what RFC w.r.t CNAMEs are you talking about? The only relevant guidelines which pop to my head are an RFC in the 1300-range or so (been a while) discussing the use of CNAMEs with MX records and such. (MX to CNAME ist verboten, but has nothing to do with web traffic)
That font argument is entertaining.
What I use GNU bc to calculate some numbers, which I then embed in a proprietary application.
Is my application now "tainted"?
IIRC, the Wavecom WISMO has been doing this for years. I have several devices based on this chip, all they need is a different user interface to be fully-functioning phones.
Hey, that may just be the solution to turning in all those papers on time..
> Knowing that it's 05:00 GMT doesn't necessarily tell you whether you're going
..works for me! :)
> to be calling a person in the middle of the night or not.
# export TZ=Canada/Eastern
# date -d "tomorrow 05:00 GMT"
Tue Aug 9 01:00:00 EDT 2005
# export TZ=US/Pacific
# !date
date -d "tomorrow 05:00 GMT"
Mon Aug 8 22:00:00 PDT 2005
> Google doesn't have magic hardware, they aren't reading through
> 800MB of your e-mail every time you do a search.
Right. They don't have magic hardware, they have (effectively) magic software. And tonnes of hardware.
It may be possible to find a good mail indexing program on the client-side -- but then I'd have to download an index huge amounts of mail every time switched workstations. Not to mention that I'd probably be stuck on a Win32 solution, since ~80% of the workstations I use are Win32... I like being able to use Macs or the ultra 5 on my desk at work for e-mail every now and then. Especially when I want to cut and paste from an active xterm into an email..
I have yet to see an IMAP server with searching which even vaguely approaches GMail's speed, although I must admit, I haven't looked in a couple of years.
Server-side side searching is a fine option, provided you have a hell of a fast server when searching *huge* amounts of mail. Which I do regularly.
I don't know how GMail does it, but I can search through ~700 MB of email in ~2s (one one thousand, two one thousand, BING!).
It would require a LOT of effort to put together a server able to do that. Hell, it would require a LOT of effort to put together a server able to read from disk that quickly.
Not that I think AJAX is the magical solution, but it's a fine solution for that app -- heavy server, light front end -- and Google has taken away the "hard part" of figuring out how to get fast searches. It used to take me in excess of 5 minutes to do what they can do in 2s. For that convenience, I will "pay the price" of having the "wrong" From: address (but I'd pay to fix that if I could...)
The Happy Hacking keyboard is too small for regular use, but it's a fine keyboard for occasional use.
:)
For a stupid solution to getting a real keyboard, run Synergy and use a Sun type 5 keyboard attached to an Ultra 5. It works.
GMail is *definately* fast for me.
But you're right, it's not faster because the network pipe is faster than my disk (duh!) -- it's faster because the back end is SO MUCH faster at searching and sorting than ANYTHING I've used on my desktop.
My typical mail usage pattern is to search through ~800MB of mail, three or four times a day, from several different boxes. The little delays when clicking "compose" and "reply" are really non-issues, compared to the time savings (over a minute per search).
What I like *BEST* about GMail is that it works (for me) as well as a desktop client (except from the From: header, GRRR!),but I can use it from *anywhere* -- and it's a DAMNED sight faster than IMAP!
The compose key is supposed to be on the right hand side of the keyboard, and have a light on it.
On a similar note, the control key is supposed to be right next to the letter 'a'.
> What would have happened if we detonated the bomb about 60 miles
> offshore of Japan in the Pacific instead?
People would've said: Ooooh, look at the pretty light! Hey, it's gettin' kinda windy, let's go inside.
It doesn't. Tell them to use another port, dick head.
You must not be a very good guru programmer if you'd leave printf() in there without any formatting arguments.
puts() would be cheaper, and the input string would be a character shorter.
> Your 1581 drive by the way is a collectable these days....
Wow, I wonder what that makes my SFD-1001s?
SCOX is down 2% (0.08) today!
Oh for the love of God, haven't you ever heard of the ServerAlias directive?
Or do you REALLY CARE that much if there happens to be a "www" in the location bar?
Download http://www.realtime-realtor.com/scripts/pngbehavio r.htc
Add IMG { behavior: url(/path/to/pngbehavior.htc); } to your CSS.
IE now supports transparent PNGs (but only for image tags). Remember to include height/width attributes when you're done developing, or it'll be a little flaky ('specially when image is already cached).
Pssst. This article has nothing to do with blogs! (ass hat)
If it makes you feel any better, I write fairly DOM/CSS/Javascript heavy web apps, and test on IE5.5 and Firefox 1.0.
It's not hard, and I don't do *any* UA detection*. The key is picking the right standards subsections, and implementing the missing stuff yourself. IE has been "good enough" since version 5.5, assuming your layout won't get broken due to text in a DIV being "off" by a few pixels, etc. Remember, users come for content, anyhow!
Incidentally, most of my stuff works on IE4.0, but it's pretty damned ugly.
* UA detection is used to pop up an alert box bitching about the browser version and recommending firefox. Does not otherwise impact site. Non-JS users get bitched at via NOSCRIPT tags.
I don't know jack about developing under Windows, but cURL is a damn fine access library for the Unix folks. Perhaps you should look into it, it probably runs under windows (bsdsockets and wsock32 are similar enough to make porting trivial if it doesn't).
Not to put too fine a point on it, but I've been serving up gzipped content for a REALLY long time... let's see... I remember that the app I originally deployed it for was running under NS4/IE4, and those were the latest browsers at the time...
IIRC, NS and IE used different Content-Encoding mime types. I think one was "x-gzip" and the other was "gzip". But really, this has been supported *forever*. I don't understand why something like mod_gzip isn't built right into Apache (1.3 -- haven't changed to 2.x yet, so if it's there, I wouldn't know).