I'm not sure why Bruce Sterling's content-free driveby merits notice as "stuff that matters". Unless perhaps the category was "Yeah, it hurts when your prejudices run into harsh reality, doesn't it?"
Why do people keep saying SPF is based on IP addresses? Here's my SPF string: "v=spf1 a mx a:safe.acme.com a:widget.acme.com a:pill.acme.com -all" Do you see any IP addresses in there? I don't. SPF is based on domain names.
And another thing. People keep complaining that SPF and other similar schemes won't stop spam cause spammers use hijacked machines. Duh! What these schemes will stop is worms, not spam. That also explains why Microsoft is interested - rather than fixing their god damned software so it's secure, they want to fix everyone's email so that broken Microsoft software isn't quite so annoying. Well, whatever.
Yeah, I haven't found a good SPF implementation either. Perl, ugh, I sure don't want that in my mail delivery path. I might just write my own C SPF milter, I've been playing with milters lately and they're a lot of fun. Here's my latest, it does IP-based blacklisting very efficiently.
By the time I woke up this morning I had already gotten more then ten thousand copies of the virus, not even counting all the bounces and stoopid virus warnings. The price of having a popular web site I guess (acme.com). I don't want to spend the CPU cycles to examine all those gigabytes, so I'm just junking any message bigger than 50KB:
Remember "red mercury"? While it was never established exactly what the stuff was, or even if it existed at all, one of the guesses was that it was exactly what this article talks about.
The energy density of these nuclear isomers is just about halfway, on a logorithmic scale, between chemical and nuclear processes. 1000 times more power than chemicals, 1000 times less powerful than fission/fusion. Approximately.
The really interesting possibility is what happens if you form some of the stuff into a rod and then start the emission at one end. Does it cascade out the other end like a gamma-ray laser? In theory, it ought to.
It's not always practical to build a tree in-memory.
Yes. When I was handed an XML problem my reaction was the same. SAX is inconvenient and DOM is too slow for large files. However instead of whinging about it, I wrote my own parser API, which doesn't fit into either of those categories. It's very simple to use, and fairly fast. Basically, it's an iterator which returns an object for each span of text in the file. The spans come with a stack of attributes showing what XML elements they are enclosed in. That's about it. I hope to release it as freeware soon. I still need a catchy three-letter acronym for it, tho.
Yeah.:-) However, even I admit that if Netcraft counted by IP address instead of by virtual hostname, thttpd's share would be much lower. There are a few huge ISPs that run it - a single thttpd process on a single CPU serving tens of thousands of hostnames. This is nice, but it's not really comparable to Apache's millions of real sites.
It's not new. As the linked article says, it's been around since 1995, so it's almost as old as the web.
>and little known
I link to it from the thttpd page, which gets a quarter million hits per month, so I guess I've done my part to make Boa better known.
>it is only CGI-based.
I don't even know what this means. It serves files using select() and non-blocking I/O, not CGI. It implements CGI for external programs. Perhaps the author meant that CGI was the only option for external programs.
>preliminary benchmarks against Apache and thttpd
As others mentioned, no, there are no Apache or thttpd benchmarks in the article. No doubt it's much faster than Apache and about as fast as thttpd, but experience shows that very few people care about web server speed.
>Boa (running under Linux)
I keep telling people that if they are one of the few who care about performance enough to run a specialized NBIO web server like thttpd or Boa or mathopd, they shouldn't throw away half their performance by running Linux. Use FreeBSD.
My thread-creation benchmark can create 100,000 threads in 10 seconds on my 800MHz Linux machine at work. No idea what kernel it's running, but I'm sure it's not recent. Furthermore, my 450MHz home machine running FreeBSD 4.5 runs the same benchmark in only five seconds. WTF?
DDR-II is basically equivalent to on-chip interleaving - the data path is clocked the same but twice as wide. Well, board makers are now doing motherboard-side DDR interleaving themselves - see recent boards based on the ServerWorks GC-LE or Intel E7500 chipsets. Some manufacturers might just decide not to bother with DDR-II.
Thanks for posting this. I never got into playing nettrek much, but did play original xtrek quite a bit. By the way, Conquest did eventually get ported to Unix in 1999, by Jon Trulson. I've played it, and it's a very close rendition of the VMS version. You can fetch it here.
I've wanted a terabyte of storage since the mid-1970s, when I realized that there were approximately a trillion square meters on the Earth's surface. Store one byte of grayscale image for each square meter and that's a terabyte of data right there.
Of course these days I'd want 3TB so I could store color images.
Slashdot Mirror
I'm not sure why Bruce Sterling's content-free driveby merits notice as "stuff that matters". Unless perhaps the category was "Yeah, it hurts when your prejudices run into harsh reality, doesn't it?"
A friend of mine advocates a multi-pronged approach to dealing with spammers and virus writers: a baseball bat with nails driven through it.
Personally I think just executing Bill Gates and liquidating Microsoft would be sufficient.
Why do people keep saying SPF is based on IP addresses? Here's my SPF string: "v=spf1 a mx a:safe.acme.com a:widget.acme.com a:pill.acme.com -all" Do you see any IP addresses in there? I don't. SPF is based on domain names.
And another thing. People keep complaining that SPF and other similar schemes won't stop spam cause spammers use hijacked machines. Duh! What these schemes will stop is worms, not spam. That also explains why Microsoft is interested - rather than fixing their god damned software so it's secure, they want to fix everyone's email so that broken Microsoft software isn't quite so annoying. Well, whatever.
Yeah, I haven't found a good SPF implementation either. Perl, ugh, I sure don't want that in my mail delivery path. I might just write my own C SPF milter, I've been playing with milters lately and they're a lot of fun. Here's my latest, it does IP-based blacklisting very efficiently.
Blacklists are for chumps.
Yes, there were many different claims about what red mercury was.
Spaceflight - I guess, although if we really want to go we'll just use fusion bombs. The dirt/energy ratio is likely to be much better.
Batteries - no, since the energy comes out as gamma rays. No known way to turn those to energy without tons of shielding / heat exchanger.
Hydrogen economy - think of it as a synthesized methanol economy and it makes more sense.
Remember "red mercury"? While it was never established exactly what the stuff was, or even if it existed at all, one of the guesses was that it was exactly what this article talks about.
The energy density of these nuclear isomers is just about halfway, on a logorithmic scale, between chemical and nuclear processes. 1000 times more power than chemicals, 1000 times less powerful than fission/fusion. Approximately.
The really interesting possibility is what happens if you form some of the stuff into a rod and then start the emission at one end. Does it cascade out the other end like a gamma-ray laser? In theory, it ought to.
...but I have it configured to look like uwm.
Pronounced 'zip'. That's pretty good.
Yes. When I was handed an XML problem my reaction was the same. SAX is inconvenient and DOM is too slow for large files. However instead of whinging about it, I wrote my own parser API, which doesn't fit into either of those categories. It's very simple to use, and fairly fast. Basically, it's an iterator which returns an object for each span of text in the file. The spans come with a stack of attributes showing what XML elements they are enclosed in. That's about it. I hope to release it as freeware soon. I still need a catchy three-letter acronym for it, tho.
Far superior program.
Yeah. :-) However, even I admit that if Netcraft counted by IP address instead of by virtual hostname, thttpd's share would be much lower. There are a few huge ISPs that run it - a single thttpd process on a single CPU serving tens of thousands of hostnames. This is nice, but it's not really comparable to Apache's millions of real sites.
>There is a pretty new
It's not new. As the linked article says, it's been around since 1995, so it's almost as old as the web.
>and little known
I link to it from the thttpd page, which gets a quarter million hits per month, so I guess I've done my part to make Boa better known.
>it is only CGI-based.
I don't even know what this means. It serves files using select() and non-blocking I/O, not CGI. It implements CGI for external programs. Perhaps the author meant that CGI was the only option for external programs.
>preliminary benchmarks against Apache and thttpd
As others mentioned, no, there are no Apache or thttpd benchmarks in the article. No doubt it's much faster than Apache and about as fast as thttpd, but experience shows that very few people care about web server speed.
>Boa (running under Linux)
I keep telling people that if they are one of the few who care about performance enough to run a specialized NBIO web server like thttpd or Boa or mathopd, they shouldn't throw away half their performance by running Linux. Use FreeBSD.
My thread-creation benchmark can create 100,000 threads in 10 seconds on my 800MHz Linux machine at work. No idea what kernel it's running, but I'm sure it's not recent. Furthermore, my 450MHz home machine running FreeBSD 4.5 runs the same benchmark in only five seconds. WTF?
I heard that 118 was going to be renamed from ununoctium to concoctium.
>compared to thttpd wich while tiny has no real performance
Buh?
Perhaps you are talking about some other web server called thttpd.
Also the i845G and P4X333 chipsets. Check the ACME Boardfinder.
DDR-II is basically equivalent to on-chip interleaving - the data path is clocked the same but twice as wide. Well, board makers are now doing motherboard-side DDR interleaving themselves - see recent boards based on the ServerWorks GC-LE or Intel E7500 chipsets. Some manufacturers might just decide not to bother with DDR-II.
The data lines are still clocked at DDR, but there are twice as many of them.
Thanks for posting this. I never got into playing nettrek much, but did play original xtrek quite a bit. By the way, Conquest did eventually get ported to Unix in 1999, by Jon Trulson. I've played it, and it's a very close rendition of the VMS version. You can fetch it here.
Same here.
Until I saw it in my referer logs this morning. Neato.
I've wanted a terabyte of storage since the mid-1970s, when I realized that there were approximately a trillion square meters on the Earth's surface. Store one byte of grayscale image for each square meter and that's a terabyte of data right there.
Of course these days I'd want 3TB so I could store color images.