Well, nice try. But you're wrong. Do a bit more research.
There are custom motherboards involved, custom low-level software involved, and I believe custom chipsets. Stratus figured that maybe they could cut their+customer's costs by using commodity CPUs (and memory and peripherals, etc) back 5-10 years ago. I'm not sure it ever sold that well because people buying million-dollar servers don't benefit that much from shaving a few thousand off the CPU. But the binary compatibility is nice.
Also, with Itanium and Itanium 2, there is support for a lot of this fancy hardware error recovery stuff in Intel-based chipsets for the first time (see the E8870 chipset docs and look for "Advanced Platform RASUM".)
Or in short, all those things that make an enterprise server an enterprise server.
Well, I sort of agree with you. However, Sun/HP/IBM were calling their Unix offerings five years ago 'enterprise' without having any of those features (even though the mainframe mostly did). I've never seen a really firm definition, although I certainly have my own views about what the phrase should mean. While I agree a bit with your point, it's also not quite fair for the 'enterprise' guys to constantly redefine the enterprise feature set to include whatever the low-end guys don't have.
You may of course disagree. The important thing is recognizing what Red Hat's enterprise solution does and does not provide.
The real question to me is, do Red Hat's 'enterprise' enhancements effectively help Linux extend dominance beyond the web-server niche which Microsoft can, should, and will try to position it into. (Promptly before Microsoft offers a low-cost version of NT server with IIS-only.)
Journal filesystems hitting maturity, logical volume management, asynchronous I/O for the database guys, TPC-C benchmarks (unaudited though?), improved clustering
There are still things Linux lacks (last I checked) that the conventional UNIX vendors have added to their systems over the last five years: things like hot-swap memory, hot-swap CPUs, memory failure resiliency (OS quits using memory if recoverable but warning-sign single-bit ECC memory errors get too great), kernel hot-patching, multipath IO, workload management stuff, and ever-more SMP/NUMA scalability.
Still, seems like Red Hat is making great strides. Hat's off! (ugh, sorry about that, couldn't resist.;-)
I've admired Dr. Tegmark's home page since he was a grad student, not so much for the design skills (ha!) but as an exemplar of mixing serious and non-serious publications for other colleauges and onlookers to enjoy, explore, and learn from. Tegmark gets the web. As for the science, some of it I can actually understand.
I would also commend to the curious Slashdot reader a couple items I found facinating from the 'non-serious' section of his website:
I've admired Dr. Tegmark's home page since he was a grad student, not so much for the design skills (ha!) but as an exemplar of mixing serious and non-serious publications for other colleauges and onlookers to enjoy, explore, and learn from. Tegmark gets the web. As for the science, some of it I can actually understand.
At the risk of straying off-topic (hey, I gave you on-topic stuff above), I would also commend to the curious Slashdot reader a couple items I found facinating from the 'non-serious'section of his website:
The funniest part of that Wired article for me was the last sentence:
A Google search on the terms "accountability," "Internet" and "pornography" brought up dozens of links, mostly to religious sites.
OK, yeah? And?
I mean, its one thing for the reporter to use Google to do a little research, but it's funny (and a bit shocking) to see one A) admit it publicly, and B) to lazily avoid actually doing the reporter's job of synthesizing the resulting material into something usefully informative for the reader.
I pretty much agree; I've never been too impressed by the chasing-taillights feeling I get with Unixes and the user interface. The commercial Unixes gave up on the desktop around 1993 when NT showed up on the scene, and while there was a resurgence of interest and improvement in the desktop with Linux vendors, there's still a long way to go. I've always been amazed at how simple things (like changing one's screen resolution) end up being hard to figure out how to do until you memorize various pseudo-obscure steps.
Still, I think you are selling short the contribution of UNIX in a couple areas at least: the web and XML.
Unix tools for accessing information have improved dramatically over time culminating in the web. I consider this a major advance, separate from the GUI. While we could argue to what extent it was a Unix vs. NeXT vs. Windows innovation (certainly it has since been coopted by Windows), it was very much a UNIX phenomena when I opened my first web browser Xmosaic back in 1993, and I think Unix deserves some credit for giving birth to it... the development linkages between TCP/IP protocols, the servers and clients on either end, and Unix are deep.
And as for XML, it seems to me to be as Unix-y an approach to the web as possible, taking Unix pipes to their logical extension: taking human readable/editable flat files and passing them between programs to manipulate data in a flexible and powerful way. As a user and a developer, I still don't think the promise of XML has fully been unleashed (it takes a lot more than a little | symbol...), but I do think its buried in there somewhere.
When asking "who is going to win this case", a key question we might see answered before the full trial would be "whose side is Intel (i.e. their developers, experts) going to testify on"?
Remember that standard used for binary compatibility of UNIX on Intel used for running SCO apps on Linux and vice-versa? iBCS2? The "Intel Binary Compatibility Standard 2nd Edition"? What do the makers of that standard have to say about this case?
What does SCO know about running UNIX on Intel that Intel doesn't know and couldn't have legally passed such knowledge to IBM or Linux communities directly?
At least some of the claims just don't pass the sniff test.
When AIX came out in 1990 or so, it was so different from SVR4 that competitors cried "AIX isn't UNIX". IBM really did a huge amount of kernel work and had all sorts of administrative features borrowed conceptually from the mainframe that were lacking in UNIX at the time. Journalling filesystems, logical volume management, etc. just for starters. The toolset for managing AIX was (and still is, to some extent) quite different from other UNIXes. I think IBM kept the SVR3/4 APIs but re-wrote most of the code below it back in the late 80s.
You know you have a weak argument when you start seeing claims like this: Added Sontag [SCO SVP], "When they (IBM) started utilizing the same engineers that worked on the Unix System V source code and the ultimate derivative of it in the form of AIX, they have effectively been applying our methods and concepts, even if there isn't a single explicit line of code" that shows up in Linux.
Believe me, there are a lot of methods and concepts that UNIX (System V/VII,SVR4) stole from predecessors, and continues to steal! Where's the suit from the MULTICS guys against AT&T? Citing executive's handwavings generalizations about leveraging UNIX expertise makes for a pretty weak legal case (and PR case imho). Monterey always had too much hand-waving for me to get enthusiastic about.
However, the issue of IBM allegedly re-creating versions of SCO libraries using SCO's actual code sounds interesting. SCO binary compatibility is actually something somewhat valuable that SCO should fully defend if it was unlawfully infringed upon via trade secret leakage rather than cleanroom reverse engineering. Still, if I remember correctly, most of the SCO/Linux binary compatibility stuff was done by some guy at SCO (Avi Tevananian or something like that? I'm butchering his name, sorry.) SCO was working on Linux binary compatibility before IBM even knew Linux existed. So to speak.
As far as infringing on how to run Linux on Intel, clearly Linus managed that without anyone's help. Now running Linux on IA-64, or perhaps some particular acceleration technique used by SCO might be grounds for a decent case depending on the particular evidences involved. I guess we'll see.
With revenues less than Red Hat, and a business model going nowhere, a legal approach makes sense for SCO. IBM can afford the lawsuit and probably route around it better than, say, Intel could route around Clipper chip infringements in existing products. If it lays clearer guidelines for Linux IP, so much the better. I guess that's what I'm hoping for. As a betting man though, I wouldn't bet on SCO for this case. Unless there are more rabbits in that hat.
I didn't subscribe at first because I hadn't gotten PayPal working. I wanted to support Slashdot, but I didn't really care about ads and was, I admit, lazy. I subscribed today... the clincher for me actually was not the early-warning, but the "More Comments..." link that lets me read earlier posts.
I've been annoyed many times how impossible it is to find an old Slashdot post I'd written on a subject. One that I couldn't get via Slashdot's (kinda weak) search feature, or even, to my great surprise, via Google. (I only found 6 or so Google links containing my username, despite having a few hundred posts over the last several years here.)
Anyway, I thought I should point out that "More Comments" link and mention it publicly lest CmdrTaco think I was really liking this new 10-minutes early thing. I don't mind it. But I really joined because I was reminded about subscribing. I've never been a 'read-every-story-when-it-appears' reader and I doubt I'll start now.
If you don't allow people to comment before a story is posted (which I agree may be wise), it'd be nice to allow users to somehow signal to you via a checkbox or something that a story is a dup.
If you want to read other's story rejects or post your own in a central place for others to view/moderate, click here.
It's a place where, as users/posters, we can experiment with an publicly-viewable, user-moderated story queue without requiring any code-changes or explicit radical business-model-switching buy-in from CmdrTaco.
My prediction is that MS will continue, for the next 2-3 versions anyway, to support legacy OpenGL 1.3 on their high-end "Professional" workstation OS versions, to support applications like those of your company, but will drop support for it on the low-end. Oh, and they probably won't ever support OpenGL 2.0 (unless ISVs like you all really bitch about it and force them to reverse course a bit.)
Um, you are missing the point. If Microsoft only puts OpenGL on the successors to Windows XP Professional , but not on the successors to Windows XP Home, it will still meet the demands of the workstation market while killing off the "volume" base of OpenGL applications.
Don't be so afraid of other people's opinions of you... after you leave that school in 3-4 years, you will never again see 99% of them; trust me, I've been there. If there's a stand that should be taken, take it. If there's a girl you want to know better, let her know!
Enlighten me. Why did the shuttle have to be going at Mach 20 and heading down at what I read in one slashdot post (take your grain of salt) was a 70% angle? Couldn't the shuttle bleed off speed by going through thin layers of atmosphere for a longer period of time leading to longer exposure to friction-generated heat but less-intense heat?
This might not be worth it due to schedule constraints in ordinary circumstances, but might be worth it in extraordinary situations.
I figure NASA guys have spent 100x as much thought as I have on the question, but my ignorance is why I'm asking. Perhaps you can enlighten me which physics constraints are involved?
[3,000 people getting killed in the WTC and the Pentagon getting attacked probably has greater relevance for the average American than even 3,000,000 people dieing to famine]
And you wonder why the rest of the world can't stand Americans.
No arrogance intended; perhaps I failed to make my point clearly. Let me try again. For someone in Africa, 3,000 people dieing in their own country is probably more relevant than 3,000,000 people dieing in America.
For the most part, the media shows what people watch. The original poster commented that this indicated people were insensitive to injustice. I did not discount that, but also pointed out that there was additionally degree of pragmatism involved. I do confess that both pragmatism and idealism are sometimes contradictory American traits.
Nice comments. Everything you say seems true but I'm still left with one question.
When I heard a NASA non-PR-official speaking earlier today, they said that even if the foam had damaged the tiles on liftoff, there was nothing they could have done about it anyway. I found the resignation troubling, albeit perhaps a bit logical. You said if they discovered the problem during reentry maneuvers, it was still too late, but he implied that even if they discovered such a problem during lift-off, it would still be too late.
My question is this: if, say, they knew they had a tile problem (not that they did in this case), couldn't they in special circumstances change their reentry trajectory so it would be more gradual in some way so that the tiles would take less heat?
To be brutally pragmatic, 3,000 people getting killed in the WTC and the Pentagon getting attacked probably has greater relevance for the average American than even 3,000,000 people dieing to famine and intra-tribal warfare off in Africa. My neighbor had 3 former coworkers die on 9/11.
It's not numbness to injustice, although that may be true. It's sheer pragmatism- the enemy of idealism perhaps but not necessarily the enemy of wisdom.
(Plus there is the news-vs-business-as-usual aspect you mention. If you want publications focusing on justice issues, not just "new"s, try donating to various charities dealing with the injustice. I have. Believe me, you'll get more information on such topics than you have time to read._
--LP
P.S. While it is worth remembering that media will most-likely show you what will help their advertisers, it's pretty well documented that various media outlets lost serious money with their 9/11 coverage due to a lack of advertising in the immediate aftermath. I'm sure though that the moneymen viewed it as a necessary investment in 'credibility', ironically viewing it as a 'justice' issue akin to the ones you feel get inadequate coverage.
Slashdot Mirror
I call bullshit and marketing.
Well, nice try. But you're wrong. Do a bit more research.
There are custom motherboards involved, custom low-level software involved, and I believe custom chipsets. Stratus figured that maybe they could cut their+customer's costs by using commodity CPUs (and memory and peripherals, etc) back 5-10 years ago. I'm not sure it ever sold that well because people buying million-dollar servers don't benefit that much from shaving a few thousand off the CPU. But the binary compatibility is nice.
Also, with Itanium and Itanium 2, there is support for a lot of this fancy hardware error recovery stuff in Intel-based chipsets for the first time (see the E8870 chipset docs and look for "Advanced Platform RASUM".)
--LP
Or in short, all those things that make an enterprise server an enterprise server.
Well, I sort of agree with you. However, Sun/HP/IBM were calling their Unix offerings five years ago 'enterprise' without having any of those features (even though the mainframe mostly did). I've never seen a really firm definition, although I certainly have my own views about what the phrase should mean. While I agree a bit with your point, it's also not quite fair for the 'enterprise' guys to constantly redefine the enterprise feature set to include whatever the low-end guys don't have.
You may of course disagree. The important thing is recognizing what Red Hat's enterprise solution does and does not provide.
The real question to me is, do Red Hat's 'enterprise' enhancements effectively help Linux extend dominance beyond the web-server niche which Microsoft can, should, and will try to position it into. (Promptly before Microsoft offers a low-cost version of NT server with IIS-only.)
--LP
Red Hat has definitely been inching up the scale.
;-)
Journal filesystems hitting maturity, logical volume management, asynchronous I/O for the database guys, TPC-C benchmarks (unaudited though?), improved clustering
There are still things Linux lacks (last I checked) that the conventional UNIX vendors have added to their systems over the last five years: things like hot-swap memory, hot-swap CPUs, memory failure resiliency (OS quits using memory if recoverable but warning-sign single-bit ECC memory errors get too great), kernel hot-patching, multipath IO, workload management stuff, and ever-more SMP/NUMA scalability.
Still, seems like Red Hat is making great strides. Hat's off! (ugh, sorry about that, couldn't resist.
--LP
I've admired Dr. Tegmark's home page since he was a grad student, not so much for the design skills (ha!) but as an exemplar of mixing serious and non-serious publications for other colleauges and onlookers to enjoy, explore, and learn from. Tegmark gets the web. As for the science, some of it I can actually understand.
I would also commend to the curious Slashdot reader a couple items I found facinating from the 'non-serious' section of his website:
a very cool diagram of "Relationships between various basic mathematical structures" from his Theory of Everything paper
and another paper addressing the question: Why does the universe have 3 spatial and 1 time dimension?
--LP
I've admired Dr. Tegmark's home page since he was a grad student, not so much for the design skills (ha!) but as an exemplar of mixing serious and non-serious publications for other colleauges and onlookers to enjoy, explore, and learn from. Tegmark gets the web. As for the science, some of it I can actually understand.
At the risk of straying off-topic (hey, I gave you on-topic stuff above), I would also commend to the curious Slashdot reader a couple items I found facinating from the 'non-serious'section of his website:
a very cool diagram of "Relationships between various basic mathematical structures" from his Theory of Everything paper
and another paper addressing the question: Why does the universe have 3 spatial and 1 time dimension?
--LP
The funniest part of that Wired article for me was the last sentence:
A Google search on the terms "accountability," "Internet" and "pornography" brought up dozens of links, mostly to religious sites.
OK, yeah? And?
I mean, its one thing for the reporter to use Google to do a little research, but it's funny (and a bit shocking) to see one A) admit it publicly, and B) to lazily avoid actually doing the reporter's job of synthesizing the resulting material into something usefully informative for the reader.
--LP
I pretty much agree; I've never been too impressed by the chasing-taillights feeling I get with Unixes and the user interface. The commercial Unixes gave up on the desktop around 1993 when NT showed up on the scene, and while there was a resurgence of interest and improvement in the desktop with Linux vendors, there's still a long way to go. I've always been amazed at how simple things (like changing one's screen resolution) end up being hard to figure out how to do until you memorize various pseudo-obscure steps.
Still, I think you are selling short the contribution of UNIX in a couple areas at least: the web and XML.
Unix tools for accessing information have improved dramatically over time culminating in the web. I consider this a major advance, separate from the GUI. While we could argue to what extent it was a Unix vs. NeXT vs. Windows innovation (certainly it has since been coopted by Windows), it was very much a UNIX phenomena when I opened my first web browser Xmosaic back in 1993, and I think Unix deserves some credit for giving birth to it... the development linkages between TCP/IP protocols, the servers and clients on either end, and Unix are deep.
And as for XML, it seems to me to be as Unix-y an approach to the web as possible, taking Unix pipes to their logical extension: taking human readable/editable flat files and passing them between programs to manipulate data in a flexible and powerful way. As a user and a developer, I still don't think the promise of XML has fully been unleashed (it takes a lot more than a little | symbol...), but I do think its buried in there somewhere.
--LP
When asking "who is going to win this case", a key question we might see answered before the full trial would be "whose side is Intel (i.e. their developers, experts) going to testify on"?
Remember that standard used for binary compatibility of UNIX on Intel used for running SCO apps on Linux and vice-versa? iBCS2? The "Intel Binary Compatibility Standard 2nd Edition"? What do the makers of that standard have to say about this case?
What does SCO know about running UNIX on Intel that Intel doesn't know and couldn't have legally passed such knowledge to IBM or Linux communities directly?
Inquiring minds want to know.
--LP
At least some of the claims just don't pass the sniff test.
When AIX came out in 1990 or so, it was so different from SVR4 that competitors cried "AIX isn't UNIX". IBM really did a huge amount of kernel work and had all sorts of administrative features borrowed conceptually from the mainframe that were lacking in UNIX at the time. Journalling filesystems, logical volume management, etc. just for starters. The toolset for managing AIX was (and still is, to some extent) quite different from other UNIXes. I think IBM kept the SVR3/4 APIs but re-wrote
most of the code below it back in the late 80s.
You know you have a weak argument when you start seeing claims like this:
Added Sontag [SCO SVP], "When they (IBM) started utilizing the same engineers that worked on the Unix System V source code and the ultimate derivative of it in the form of AIX, they have
effectively been applying our methods and concepts, even if there isn't a single explicit line of code" that shows up in Linux.
Believe me, there are a lot of methods and concepts that UNIX (System V/VII,SVR4) stole from predecessors, and continues to steal! Where's the suit from the MULTICS guys against AT&T? Citing executive's handwavings generalizations about leveraging UNIX expertise makes for a pretty weak legal case (and PR case imho). Monterey always had too much hand-waving for me to get enthusiastic about.
However, the issue of IBM allegedly re-creating versions of SCO libraries using SCO's actual code sounds interesting. SCO binary compatibility is actually something somewhat valuable that SCO should fully defend if it was unlawfully infringed upon via trade secret leakage rather than cleanroom reverse engineering. Still, if I remember correctly, most of the SCO/Linux binary compatibility stuff was done by some guy at SCO (Avi Tevananian or something like that? I'm butchering his name, sorry.) SCO was working on Linux binary compatibility before IBM even knew Linux existed. So to speak.
As far as infringing on how to run Linux on Intel, clearly Linus managed that without anyone's help. Now running Linux on IA-64, or perhaps some particular acceleration technique used by SCO might be grounds for a decent case depending on the particular evidences involved. I guess we'll see.
With revenues less than Red Hat, and a business model going nowhere, a legal approach makes sense for SCO. IBM can afford the lawsuit and probably route around it better than, say, Intel could route around Clipper chip infringements in existing products. If it lays clearer guidelines for Linux IP, so much the better. I guess that's what I'm hoping for. As a betting man though, I wouldn't bet on SCO for this case. Unless there are more rabbits in that hat.
--LP
I didn't subscribe at first because I hadn't gotten PayPal working. I wanted to support Slashdot, but I didn't really care about ads and was, I admit, lazy. I subscribed today... the clincher for me actually was not the early-warning, but the "More Comments..." link that lets me read earlier posts.
I've been annoyed many times how impossible it is to find an old Slashdot post I'd written on a subject. One that I couldn't get via Slashdot's (kinda weak) search feature, or even, to my great surprise, via Google. (I only found 6 or so Google links containing my username, despite having a few hundred posts over the last several years here.)
Anyway, I thought I should point out that "More Comments" link and mention it publicly lest CmdrTaco think I was really liking this new 10-minutes early thing. I don't mind it. But I really joined because I was reminded about subscribing. I've never been a 'read-every-story-when-it-appears' reader and I doubt I'll start now.
--LP
If you don't allow people to comment before a story is posted (which I agree may be wise), it'd be nice to allow users to somehow signal to you via a checkbox or something that a story is a dup.
--LP
Users are welcome to moderate/view story ideas from one another here.
It's an experimental open queue. Submit your favorite stories that got rejected...
--LP
If you want to read other's story rejects or post your own in a central place for others to view/moderate, click here.
It's a place where, as users/posters, we can experiment with an publicly-viewable, user-moderated story queue without requiring any code-changes or explicit radical business-model-switching buy-in from CmdrTaco.
--LP
If you want a user-moderated, publicly-viewable story queue, submit your rejected stories and read others here.
It's a place for a little experimentation for us users without getting CmdrTaco's explicit consent or requiring further development work.
--LP
My prediction is that MS will continue, for the next 2-3 versions anyway, to support legacy OpenGL 1.3 on their high-end "Professional" workstation OS versions, to support applications like those of your company, but will drop support for it on the low-end. Oh, and they probably won't ever support OpenGL 2.0 (unless ISVs like you all really bitch about it and force them to reverse course a bit.)
--LP
Um, you are missing the point. If Microsoft only puts OpenGL on the successors to Windows XP Professional , but not on the successors to Windows XP Home, it will still meet the demands of the workstation market while killing off the "volume" base of OpenGL applications.
--LP
Don't be so afraid of other people's opinions of you... after you leave that school in 3-4 years, you will never again see 99% of them; trust me, I've been there. If there's a stand that should be taken, take it. If there's a girl you want to know better, let her know!
--LP
And if you get rejected, submit it to the user-moderated journal queue I'm trying to get going...
--LP
It seems that you will need to buy a separate antenna with most if not all of these PC cards (or get cable HDTV with a matching connector).
Oh, and to see if there are HDTV signals in your area, try entering your zip code at The HDTV Pub.
--LP
Hypothetically, all my Slashdot posts should be archived on Google, at least the static pages from two or more weeks old.
/. says I've made 393 posts over the last three years or so.
However, Google seems to only have 6-7 webpages when I search for "LinuxParanoid", while
Any ideas why?
--LP
P.S. (Slashdot's search engine is also pragmatically worthless for me finding my old posts, which is one reason why I noticed.)
Inquiring minds want to know...
Enlighten me. Why did the shuttle have to be going at Mach 20 and heading down at what I read in one slashdot post (take your grain of salt) was a 70% angle? Couldn't the shuttle bleed off speed by going through thin layers of atmosphere for a longer period of time leading to longer exposure to friction-generated heat but less-intense heat?
This might not be worth it due to schedule constraints in ordinary circumstances, but might be worth it in extraordinary situations.
I figure NASA guys have spent 100x as much thought as I have on the question, but my ignorance is why I'm asking. Perhaps you can enlighten me which physics constraints are involved?
--LP
[3,000 people getting killed in the WTC and the Pentagon getting attacked probably has greater relevance for the average American than even 3,000,000 people dieing to famine]
And you wonder why the rest of the world can't stand Americans.
No arrogance intended; perhaps I failed to make my point clearly. Let me try again. For someone in Africa, 3,000 people dieing in their own country is probably more relevant than 3,000,000 people dieing in America.
For the most part, the media shows what people watch. The original poster commented that this indicated people were insensitive to injustice. I did not discount that, but also pointed out that there was additionally degree of pragmatism involved. I do confess that both pragmatism and idealism are sometimes contradictory American traits.
--LP
Nice comments. Everything you say seems true but I'm still left with one question.
When I heard a NASA non-PR-official speaking earlier today, they said that even if the foam had damaged the tiles on liftoff, there was nothing they could have done about it anyway. I found the resignation troubling, albeit perhaps a bit logical. You said if they discovered the problem during reentry maneuvers, it was still too late, but he implied that even if they discovered such a problem during lift-off, it would still be too late.
My question is this: if, say, they knew they had a tile problem (not that they did in this case), couldn't they in special circumstances change their reentry trajectory so it would be more gradual in some way so that the tiles would take less heat?
--LP
To be brutally pragmatic, 3,000 people getting killed in the WTC and the Pentagon getting attacked probably has greater relevance for the average American than even 3,000,000 people dieing to famine and intra-tribal warfare off in Africa. My neighbor had 3 former coworkers die on 9/11.
It's not numbness to injustice, although that may be true. It's sheer pragmatism- the enemy of idealism perhaps but not necessarily the enemy of wisdom.
(Plus there is the news-vs-business-as-usual aspect you mention. If you want publications focusing on justice issues, not just "new"s, try donating to various charities dealing with the injustice. I have. Believe me, you'll get more information on such topics than you have time to read._
--LP
P.S. While it is worth remembering that media will most-likely show you what will help their advertisers, it's pretty well documented that various media outlets lost serious money with their 9/11 coverage due to a lack of advertising in the immediate aftermath. I'm sure though that the moneymen viewed it as a necessary investment in 'credibility', ironically viewing it as a 'justice' issue akin to the ones you feel get inadequate coverage.