actually, I pin it to lousy programmers writing lousy libraries. the preferred (don't fetch it) mechanism should be the *easiest* to implement, rather than being an inconsistent case of obscurity (that the sax system created).
but really, it's 'cause most programmers actually suck at writing libraries, but their libraries get used (and thus, have backwards-compatibility maintenance issues) before they can get them refined into the right way to do things.
Caching it requires having access to get it in the first place, doesn't it? I can't ask them to open their system up to the 'net just to grab some file every time they restart their servers, can I?
See my comment elsewhere in this thread - there's some "magic" that Java can do to load a local file as a resource given an external URL but I've never seen a decent example of how to set that up for my own apps.
In general, you're right. The various standard languages built in XML do basically require and expect that you're able to look at files on the internet. XML wasn't designed for many of the private-network uses its being used for today, and neither were many of the tools that reference public DTDs.
However, with most parsers there is a way to set up a mapping where it can look on the local box for a file matching a specific URL. JBoss, Hibernate, and Struts all do that for DTD files they have in their respective Jars, loading them as resources. However, I've never seen a clear example of how to do that for other files. It's "black box magic" that I've never found a reasonable, simple example of how to implement in my own systems.
Google doesn't help because "Java" "XML" "Resource File" still brings up thousands of entries I don't have time to nail down. And trying to dig through the thousands of lines of source on those apps is just as much a search I don't have time for.
we have licensed customers with this (in fact, they're the problem - internally we don't see any problem with having outbound internet access for our servers and trust the firewalls we put up); i can't go telling their IT departments (who are sometimes completely inept) that we need to go mucking around with OS configuration files, especially in Windows.
Actually the DTD is loaded up by pretty much every proper XML library even if validation is "off".
The DTD contains more than just the element definitions and hierarchy. Its also used to define entities (&...;) that are non-standard to XML but may be expected in the file. HTML has tons of pre-defined entities but XML only has the core 4. All others are defined in DTDs and loaded on the fly as part of the processing.
There are ways to turn it off at the lowest levels, but higher-level abstractions/libraries might not give access to that. For example, with JAXP + SAX you can turn off DTD loading, but Jakarta Commons Digester doesn't give a setting where you can trigger that, so Digester tries to load the dtd, and even with validation off you can't change that. My only recourse is to take the DTD lines out of the various config files. (Reason: My JBoss server is deployed in private networks where the server can't reach the internet).
When the Doctor Who episodes on bittorrent change, I'll get a new codec. For now, that's all I use it for - getting the doctor whos from Britain 'cause i'm not willing to wait for sci-fi to catch up.
I'm still waiting on that one......of course it doesn't really matter much since iTunes is in control of the audio-for-sale and video-for-sale market these days. DIVx-based AVIs are becoming the default for "free" video content, particularly from Europe.
As I wrote before here, WMP for Linux was meant as a strategic move to scare content owners away from the open-sourcing of Real Networks' player and codecs, by promising WMP-based DRM for the Linux market. It seemed to work, but rather than go to WMP (which had technical issues as shown by early BootlegTV downloads from the DGM record label (King Crimson)), they held off until iTunes set the new DRM standard. M$'s been behind ever since.
To think that the targetted consumer of this is a "switch from Outlook Express" user is extremely short sighted and arrogant.
Thunderbird has all the tools needed (through extensions) to be the power mailer for the power user.
But that is not its intention. It was, like Firefox, intended to be an end-user tool, and you make end user tools by enforcing interface *simplicity*. It is meant to be good enough for OEMs like Dell and HP to finally stop bundling the garbage that is OE and provide a better experience. For the average user, a "better experience" means *fewer* configuration dialogs and settings, not more choices. People HATE choices. Drive down a busy highway sometime and watch closely: you'll see the worst backups every time someone needs to make a choice - it's not the incoming or outgoing traffic that screws people up and makes them slow down: its the choice. People simply don't want to have to make decisions, nor do they want to have to fill in dialogs that in turn give them choices to have to make later: the thing should "just know".
well, some SMTP configurations will allow you to send the message, regardless of the "from" setting in the message, by verifying the login. Other configurations lock you down so you can only send messages if the "from" belongs to the domain of the SMTP server itself (my work does this). It's a matter of how the SMTP admin decided to close the gateways (well, loopholes, really) that permit spam outbound.
I haven't needed to change SMTP settings for my personal mailbox in years; it allows me to send it out with different "From" emails.
However, while my wife was working an ISP that DID lock the SMTP settings in that way, the SMTP-selection feature from the thunderbird extension was a life-saver for being able to switch back and forth between the test and production servers to know that things were working.
I was about to say (having finally read the article) - *most* users only need one SMTP server, but there are extensions that make it easy to set up alternates. This is why extensions exist: keep the basic interface simple, allow "power users" to improve things.
In the infamous Secret Service seizure of Steve Jackson Games' Illuminati Online BBS system in 1990 (case resolved in 1993), the court found that the government reading unread emails on a machine by seizure of the machine was not "wire-tapping", in spite of arguments by the EFF that the end result is the same - the government sees your communication before you do.
For all of the alledged "protections" congress has given electronic communication, they've all been mere extensions of protection for variations of wire-tapping. If the government can actually get the physical hardware in their hands, anything goes. There is no sense of protected files or folders on a disk drive.
actually, it's like the current analog/digital issue with mainline phone wire carriers. or actually "pulse" vs "dialtone".
there's a $2 charge on your local phone bill (if you haven't gone VOIP) that's there to support dialtone usage.
now in actuality, it's more expensive for the phone companies to maintain the ancient pulse-based system of the 50s and 60s than it is to support the dialtone, but that wasn't the case when dialtone first showed up. So they charged extra for dialtone support.
now, we still pay extra, with some of that money going to ensure backwards compatibility to phones that simply no longer exist.
so too, the "digital" upcharge. it really is already more expensive (at a cost per mile, not cost per customer, basis - the analog system requires more power and uses up far too much bandwidth that the digital signal can compress) to maintain the analog signal than it is to just shut the whole thing down and go 100% digital. but they've already added that digital upcharge which means there will always be those last holdouts.
if they had instead sold a basic digital package exactly matching the analog package in terms of channels and uptime, they'd have changed everybody over and saved tons of money by dropping the analog signal. THEN, once everybody was changed over, assert their monopoly priviledge and raise the costs again like a good little for-profit corporation...
oh yeah - the capture of gollum took place in the 17 years between bilbo's birthday and frodo's departure, hence being so critical in the book version of the council of elrond (and the reason Legolas was there in the first place). shows what i get for not having a copy of appendix B with me at all times...
that 17 years got very short-shrifted in the film.
well, there's 60 years between the Hobbit and Bilbo's Birthday that opens LotR, plus parallels to the Hobbit (what was Gandalf doing while he wasn't on the trip? Going to Isengard, dealing with the evil down (a reformation attempt by sauron) in Dol Guldur, etc. Gandalf and Aragorn do a lot during that 60 years, including following Gollum all the way to Mordor. Then there's Balin's attempt to recolonize Moria, Sauruman's first visage into the palantir, etc...
there are lots of little bits and pieces to work from in that time-frame, with the only trouble being that none of them alone have enough Tolkien documentation on them to actually work into a script, in my opinion. just a collection of short stories...
Marillion allowed people to download (for a small fee, 'cause it was an expensive proposition) the separate tracks (and alternates), including unprocessed vocals, for their Anaraknophobia album back in 2000, and took the best remixes from the fans and released them as Remixomatosis a couple of years later.
does that include the new pressing that just came out a week ago?
or are you a griper who hates the Director's Cut and prefers the theatrical? (or like me wishes Ridley would do that magic he did with Legend and put them BOTH out, beautifully remastered, on the same set with a decent making-of like any GOOD dvd release should be?)
...and you'll know how serious any Microsoft announcement about software for Linux is.
3 or 4 years ago,/. had the announcement of WMP for Linux (which I, correctly I believe, posited that it was both vaporware and a strategic announcement to get content providers away from RealPlayer, then the only DRM system that officially supported Linux).
This one makes even less sense, as there's no target, no commercial enterprise that has a potential market for office for Linux (OO is free and if OO didn't come out, the Gnome office suite would probably have gotten more development and attention). Nobody has the potential in the Office suite to use Linux as a means of saying "we're better than Microsoft" to any content providers providing proprietary material.
So unless its going to be part of a larger "patent scare" program Microsoft might pull (they've been holding THAT trump card on Office apps for years), I don't see the point.
And if there's no point, there's no truth to it. Nothing Microsoft does it does without a specific competitor in mind, and there really is no competitor here.
Slashdot Mirror
actually, I pin it to lousy programmers writing lousy libraries. the preferred (don't fetch it) mechanism should be the *easiest* to implement, rather than being an inconsistent case of obscurity (that the sax system created).
but really, it's 'cause most programmers actually suck at writing libraries, but their libraries get used (and thus, have backwards-compatibility maintenance issues) before they can get them refined into the right way to do things.
well, as long as they're talking about an XML "2.0", maybe you should write to a committee member and submit the idea?
Caching it requires having access to get it in the first place, doesn't it? I can't ask them to open their system up to the 'net just to grab some file every time they restart their servers, can I?
See my comment elsewhere in this thread - there's some "magic" that Java can do to load a local file as a resource given an external URL but I've never seen a decent example of how to set that up for my own apps.
In general, you're right. The various standard languages built in XML do basically require and expect that you're able to look at files on the internet. XML wasn't designed for many of the private-network uses its being used for today, and neither were many of the tools that reference public DTDs.
However, with most parsers there is a way to set up a mapping where it can look on the local box for a file matching a specific URL. JBoss, Hibernate, and Struts all do that for DTD files they have in their respective Jars, loading them as resources. However, I've never seen a clear example of how to do that for other files. It's "black box magic" that I've never found a reasonable, simple example of how to implement in my own systems.
Google doesn't help because "Java" "XML" "Resource File" still brings up thousands of entries I don't have time to nail down. And trying to dig through the thousands of lines of source on those apps is just as much a search I don't have time for.
seriously, how long will it be in that general area of the sky at that time? DC's got rain and clouds for at least the next 2 days...
which "hosts file"?
we have licensed customers with this (in fact, they're the problem - internally we don't see any problem with having outbound internet access for our servers and trust the firewalls we put up); i can't go telling their IT departments (who are sometimes completely inept) that we need to go mucking around with OS configuration files, especially in Windows.
Actually the DTD is loaded up by pretty much every proper XML library even if validation is "off".
The DTD contains more than just the element definitions and hierarchy. Its also used to define entities (&...;) that are non-standard to XML but may be expected in the file. HTML has tons of pre-defined entities but XML only has the core 4. All others are defined in DTDs and loaded on the fly as part of the processing.
There are ways to turn it off at the lowest levels, but higher-level abstractions/libraries might not give access to that. For example, with JAXP + SAX you can turn off DTD loading, but Jakarta Commons Digester doesn't give a setting where you can trigger that, so Digester tries to load the dtd, and even with validation off you can't change that. My only recourse is to take the DTD lines out of the various config files. (Reason: My JBoss server is deployed in private networks where the server can't reach the internet).
When the Doctor Who episodes on bittorrent change, I'll get a new codec. For now, that's all I use it for - getting the doctor whos from Britain 'cause i'm not willing to wait for sci-fi to catch up.
I'm still waiting on that one... ...of course it doesn't really matter much since iTunes is in control of the audio-for-sale and video-for-sale market these days. DIVx-based AVIs are becoming the default for "free" video content, particularly from Europe.
As I wrote before here, WMP for Linux was meant as a strategic move to scare content owners away from the open-sourcing of Real Networks' player and codecs, by promising WMP-based DRM for the Linux market. It seemed to work, but rather than go to WMP (which had technical issues as shown by early BootlegTV downloads from the DGM record label (King Crimson)), they held off until iTunes set the new DRM standard. M$'s been behind ever since.
...to be on a mailing, phone, and spam list for telemarketers on Mars.
To think that the targetted consumer of this is a "switch from Outlook Express" user is extremely short sighted and arrogant.
Thunderbird has all the tools needed (through extensions) to be the power mailer for the power user.
But that is not its intention. It was, like Firefox, intended to be an end-user tool, and you make end user tools by enforcing interface *simplicity*. It is meant to be good enough for OEMs like Dell and HP to finally stop bundling the garbage that is OE and provide a better experience. For the average user, a "better experience" means *fewer* configuration dialogs and settings, not more choices. People HATE choices. Drive down a busy highway sometime and watch closely: you'll see the worst backups every time someone needs to make a choice - it's not the incoming or outgoing traffic that screws people up and makes them slow down: its the choice. People simply don't want to have to make decisions, nor do they want to have to fill in dialogs that in turn give them choices to have to make later: the thing should "just know".
Go read http://joelonsoftware.com/ for a while.
well, some SMTP configurations will allow you to send the message, regardless of the "from" setting in the message, by verifying the login. Other configurations lock you down so you can only send messages if the "from" belongs to the domain of the SMTP server itself (my work does this). It's a matter of how the SMTP admin decided to close the gateways (well, loopholes, really) that permit spam outbound.
I haven't needed to change SMTP settings for my personal mailbox in years; it allows me to send it out with different "From" emails.
However, while my wife was working an ISP that DID lock the SMTP settings in that way, the SMTP-selection feature from the thunderbird extension was a life-saver for being able to switch back and forth between the test and production servers to know that things were working.
I was about to say (having finally read the article) - *most* users only need one SMTP server, but there are extensions that make it easy to set up alternates. This is why extensions exist: keep the basic interface simple, allow "power users" to improve things.
In the infamous Secret Service seizure of Steve Jackson Games' Illuminati Online BBS system in 1990 (case resolved in 1993), the court found that the government reading unread emails on a machine by seizure of the machine was not "wire-tapping", in spite of arguments by the EFF that the end result is the same - the government sees your communication before you do.
For all of the alledged "protections" congress has given electronic communication, they've all been mere extensions of protection for variations of wire-tapping. If the government can actually get the physical hardware in their hands, anything goes. There is no sense of protected files or folders on a disk drive.
or the giant gas-guzzling SUV in my parking garage with the license plate "ERTH LUV".
actually, it's like the current analog/digital issue with mainline phone wire carriers. or actually "pulse" vs "dialtone".
there's a $2 charge on your local phone bill (if you haven't gone VOIP) that's there to support dialtone usage.
now in actuality, it's more expensive for the phone companies to maintain the ancient pulse-based system of the 50s and 60s than it is to support the dialtone, but that wasn't the case when dialtone first showed up. So they charged extra for dialtone support.
now, we still pay extra, with some of that money going to ensure backwards compatibility to phones that simply no longer exist.
so too, the "digital" upcharge. it really is already more expensive (at a cost per mile, not cost per customer, basis - the analog system requires more power and uses up far too much bandwidth that the digital signal can compress) to maintain the analog signal than it is to just shut the whole thing down and go 100% digital. but they've already added that digital upcharge which means there will always be those last holdouts.
if they had instead sold a basic digital package exactly matching the analog package in terms of channels and uptime, they'd have changed everybody over and saved tons of money by dropping the analog signal. THEN, once everybody was changed over, assert their monopoly priviledge and raise the costs again like a good little for-profit corporation...
save early, save often, and if you have power problems THAT bad, you must live in Bagdad.
oh yeah - the capture of gollum took place in the 17 years between bilbo's birthday and frodo's departure, hence being so critical in the book version of the council of elrond (and the reason Legolas was there in the first place). shows what i get for not having a copy of appendix B with me at all times...
that 17 years got very short-shrifted in the film.
well, there's 60 years between the Hobbit and Bilbo's Birthday that opens LotR, plus parallels to the Hobbit (what was Gandalf doing while he wasn't on the trip? Going to Isengard, dealing with the evil down (a reformation attempt by sauron) in Dol Guldur, etc. Gandalf and Aragorn do a lot during that 60 years, including following Gollum all the way to Mordor. Then there's Balin's attempt to recolonize Moria, Sauruman's first visage into the palantir, etc...
there are lots of little bits and pieces to work from in that time-frame, with the only trouble being that none of them alone have enough Tolkien documentation on them to actually work into a script, in my opinion. just a collection of short stories...
How about "has failed to understand for the last 20 years they've been saying this."
/. bothered, because this exact same statement has come from them repeatedly for as long as I've been following computers.
I'm surprised
Marillion allowed people to download (for a small fee, 'cause it was an expensive proposition) the separate tracks (and alternates), including unprocessed vocals, for their Anaraknophobia album back in 2000, and took the best remixes from the fans and released them as Remixomatosis a couple of years later.
who thinks the whole "space elevator" concept is just downright STUPID?
ah, Candleshoe. forgot about that scene. still, she got him back, which i suppose makes it kinda alright.
does that include the new pressing that just came out a week ago?
or are you a griper who hates the Director's Cut and prefers the theatrical? (or like me wishes Ridley would do that magic he did with Legend and put them BOTH out, beautifully remastered, on the same set with a decent making-of like any GOOD dvd release should be?)
...and you'll know how serious any Microsoft announcement about software for Linux is.
/. had the announcement of WMP for Linux (which I, correctly I believe, posited that it was both vaporware and a strategic announcement to get content providers away from RealPlayer, then the only DRM system that officially supported Linux).
3 or 4 years ago,
This one makes even less sense, as there's no target, no commercial enterprise that has a potential market for office for Linux (OO is free and if OO didn't come out, the Gnome office suite would probably have gotten more development and attention). Nobody has the potential in the Office suite to use Linux as a means of saying "we're better than Microsoft" to any content providers providing proprietary material.
So unless its going to be part of a larger "patent scare" program Microsoft might pull (they've been holding THAT trump card on Office apps for years), I don't see the point.
And if there's no point, there's no truth to it. Nothing Microsoft does it does without a specific competitor in mind, and there really is no competitor here.