Extensions aren't too hard. Feedparser shows the way to do it--keep a list of known namespace declaration URLs. Then when you hit an xmlns declaration, you record the prefix they specify, and the standard one you prefer, in a hash.
Once you've done that, you just need to preprocess every XML element by mapping the namespaces through the hash to standardize them. I then map the namespaced element to an object field by converting the : to an _.
With no indication of the type of the textual data in description elements, the only way to include HTML in RSS is to assume that the reader will sniff for HTML by looking at the contents and guessing the data type. Any programmer who's not an inexperienced hack will realize that's not a good idea.
Bullcrap. RSS 2.0 and RSS 1.0 have no way to indicate whether content is HTML or text, so they do not support HTML.
Sure, some people shove HTML in the text elements, and some feed readers sniff the text to see if it looks like HTML, but that's a long way from saying the format supports it.
Well, developers care, because the RSS specs (all nine different ones) are a mess, whereas Atom is quite carefully specified. RSS has the same problems as "Netscape HTML", whereas Atom is more like XHTML.
NIST's published instructions on setting up NTP for Windows and Mac users actively invite using them as your NTP server. (Check the PDFs.) If they don't want to be used as time server, they shouldn't have invited it.
As the reviewer said, precise synchronization is absolutely critical for forensic purposes. Can you conclusively demonstrate that Event A on machine Foo occurred before Event B on machine Bar?
If you think you might ever need to do that, better hope you're not using BSD syslog or that the intruders are kind enough not to do so during the wrong time window.
The point being that a psycopathic manager, while many of the characteristics may seem desirable in isolation, is there only to serve his/her own goals and has no loyalty to the company. They do what they do for their own self interests, and if there are shortcuts that help them achieve it they will take those shortcuts even if it harms the company.
If you look at the next level up, that's exactly how the company behaves towards society. In fact, that's how the company is required to behave. Oh, the irony.
A virus could easily be extremely malicious, yet unlikely to be detected for days.
For example, it would be relatively simple to write a virus which had a database of common names, and rude words to replace them with. It would know enough about Word's file format to seek out Word documents and quietly switch the names. You could do the same with web pages.
Most businesses wouldn't notice, until someone sent a letter to a major client starting "Dear Dick Head..." or the press wondered why the CEO's web page called him "Fat Crook".
You could even make the substitute words the same length as the search words, so you wouldn't need to understand the file format and wouldn't need to rewrite files. Target the newest files first for maximum effect, or target the oldest files first for longest time before detection. You might even manage to hold out undetected for long enough that people's backups would be corrupted too.
If we're re-introducing extinct fauna to America, why don't we try re-introducing liberals?
As recently as the 1930s the liberal was found in great herds across the continent, even throughout Texas. This good-natured creature was sadly unable to adapt to defend itself against a new species of vicious and ruthless predator. Nowadays the liberal is all but extinct, surviving only in a few isolated colonies such as Berkeley, Cambridge and Austin.
It seems like you're reading a different Perl 6 design commentary from me. To me, Apocalypse 12 (for example) is a cruft sandwich with a side-order of bloat, sprinkled with liberally with sigils. More thoughts at http://www.xciv.org/~meta/2005/03/02#2005-03-02, but basically the way I see it is that Perl doesn't need more ideas, it needs fewer.
Almost all video game conversions of movies (and TV shows) fucking suck, just like all movie versions of video games fucking suck.
Looking at the Metacritic list of PS2 games in score order, the best movie game ever is Lord of the Rings: Return of the King, at 85%--but that puts it at #183 on the chart. So basically, there are hundreds of games better than the very best video game ever based on a movie.
Now look at the bottom of the chart. By my count, 15 of the 40 worst PS2 games ever are movie/TV games.
Personally, I find it amazing that people are still willing to throw money at developing video game tie-ins for movies and TV, and even more amazing that suckers are willing to buy them.
It always amazes me when SCEE creates a game, already in English, and SCEA passes up releasing it in the US. We eventually got Wipeout Fusion and Dropship via a third party distributor.
The "copyright infringement isn't theft" is my favorite, as it in no way justifies breaking of the law.
Perhaps eventually, then, the proponents of copyright will wise up and stop lying that copyright infringement is theft. That way there will be no need for distracting "Copyright infringement isn't theft" responses, yes?
I still haven't seen a PSP. Until I see one, I'm not likely to want it, am I?
Come on, Sony, get that clue you so desperately need. Start having PSPs on display in stores. I know this will sound strange, but people tend not to buy $300 gadgets sight unseen. Maybe they did in the.com 90s, but not any more.
Slashdot Mirror
If Google use Jabber, users of the Google service will be able to message anyone on AIM, and anyone on MSN.
Extensions aren't too hard. Feedparser shows the way to do it--keep a list of known namespace declaration URLs. Then when you hit an xmlns declaration, you record the prefix they specify, and the standard one you prefer, in a hash.
Once you've done that, you just need to preprocess every XML element by mapping the namespaces through the hash to standardize them. I then map the namespaced element to an object field by converting the : to an _.
The problem is how you tell whether something is encoded HTML or not. As I've said elsewhere, sniffing the content is not a good solution.
With no indication of the type of the textual data in description elements, the only way to include HTML in RSS is to assume that the reader will sniff for HTML by looking at the contents and guessing the data type. Any programmer who's not an inexperienced hack will realize that's not a good idea.
http://www.w3.org/2005/Atom
By implementing parsers for both.
Bullcrap. RSS 2.0 and RSS 1.0 have no way to indicate whether content is HTML or text, so they do not support HTML.
Sure, some people shove HTML in the text elements, and some feed readers sniff the text to see if it looks like HTML, but that's a long way from saying the format supports it.
Someone has to implement the feed parser. Those people, i.e. developers, care.
Well, developers care, because the RSS specs (all nine different ones) are a mess, whereas Atom is quite carefully specified. RSS has the same problems as "Netscape HTML", whereas Atom is more like XHTML.
RSS has 11 different varieties, 9 if you exclude the two attempts at an "RSS 3". Atom has a single variety.
RSS 1.0 has a way to include HTML in the feed. RSS 2.0 doesn't. Atom does, and also supports XHTML.
RSS 1.0 is extensible in a standard way via namespaces. RSS 2.0 is extended via ad-hoc additions. Atom is extensible via namespaces.
Atom is more complicated than RSS 1.0, which is more complicated than RSS 2.0.
According to chronyd's documentation, it doesn't support leap seconds. WTF is the point of an NTP replacement that doesn't handle leap seconds?!
NIST's published instructions on setting up NTP for Windows and Mac users actively invite using them as your NTP server. (Check the PDFs.) If they don't want to be used as time server, they shouldn't have invited it.
If you think you might ever need to do that, better hope you're not using BSD syslog or that the intruders are kind enough not to do so during the wrong time window.
Do you really need to ask?
If you look at the next level up, that's exactly how the company behaves towards society. In fact, that's how the company is required to behave. Oh, the irony.
A virus could easily be extremely malicious, yet unlikely to be detected for days.
For example, it would be relatively simple to write a virus which had a database of common names, and rude words to replace them with. It would know enough about Word's file format to seek out Word documents and quietly switch the names. You could do the same with web pages.
Most businesses wouldn't notice, until someone sent a letter to a major client starting "Dear Dick Head..." or the press wondered why the CEO's web page called him "Fat Crook".
You could even make the substitute words the same length as the search words, so you wouldn't need to understand the file format and wouldn't need to rewrite files. Target the newest files first for maximum effect, or target the oldest files first for longest time before detection. You might even manage to hold out undetected for long enough that people's backups would be corrupted too.
If we're re-introducing extinct fauna to America, why don't we try re-introducing liberals?
As recently as the 1930s the liberal was found in great herds across the continent, even throughout Texas. This good-natured creature was sadly unable to adapt to defend itself against a new species of vicious and ruthless predator. Nowadays the liberal is all but extinct, surviving only in a few isolated colonies such as Berkeley, Cambridge and Austin.
I basically don't like syntax at all, which is why I like Ruby. There's very little syntax to remember, and it's all remarkably consistent.
I gave up Perl this year. Now I write Ruby. Never been happier. Ruby is what Perl 6 ought to be.
It seems like you're reading a different Perl 6 design commentary from me. To me, Apocalypse 12 (for example) is a cruft sandwich with a side-order of bloat, sprinkled with liberally with sigils. More thoughts at http://www.xciv.org/~meta/2005/03/02#2005-03-02, but basically the way I see it is that Perl doesn't need more ideas, it needs fewer.
(Still writing Ruby, still liking it.)
I used Solaris 2 for years and loved it.
However, in the wake of the SCO lawsuit, why would anyone in their right mind touch anything tainted with System V code?
Almost all video game conversions of movies (and TV shows) fucking suck, just like all movie versions of video games fucking suck.
Looking at the Metacritic list of PS2 games in score order, the best movie game ever is Lord of the Rings: Return of the King, at 85%--but that puts it at #183 on the chart. So basically, there are hundreds of games better than the very best video game ever based on a movie.
Now look at the bottom of the chart. By my count, 15 of the 40 worst PS2 games ever are movie/TV games.
Personally, I find it amazing that people are still willing to throw money at developing video game tie-ins for movies and TV, and even more amazing that suckers are willing to buy them.
Probably. Wouldn't be the first time.
It always amazes me when SCEE creates a game, already in English, and SCEA passes up releasing it in the US. We eventually got Wipeout Fusion and Dropship via a third party distributor.
Perhaps eventually, then, the proponents of copyright will wise up and stop lying that copyright infringement is theft. That way there will be no need for distracting "Copyright infringement isn't theft" responses, yes?
I still haven't seen a PSP. Until I see one, I'm not likely to want it, am I?
.com 90s, but not any more.
Come on, Sony, get that clue you so desperately need. Start having PSPs on display in stores. I know this will sound strange, but people tend not to buy $300 gadgets sight unseen. Maybe they did in the