Posts by James Butler 329 publicly visible posts • joined 17 Aug 2007
If all you're interested in is games, then get a game system and leave comments about computers to those of us who know what they are supposed to be used for. As evidence of your ignorance, I point out that games function on the platform for which they are written. That has nothing to do with the OS and everything to do with the developers.
Get a PlayStation or, hell, even an XBox. You'll be happier and seem smarter.
We are heading back to SGML, from which HTML was but a tiny subset. The notion of publisher-definable syntax and document interoperability have been around and in use for years before Tim B-L released HTML. HTML set us back because "the masses" needed to be able to publish at will using stupidly simple concepts.
Now we are seeing that "new" ideas are running into the same wall that SGML ran into .. client capabilities. It does no good if you specify all of your syntax to the nth degree if the client is so stupid it doesn't bother to follow your instructions (DTD). There is nothing more frustrating than perfecting your schemas only to find that the vast majority of people can't view the results of its implementation because the browser they use is incredibly limited in its parsing and rendering abilities.
Let's get back to browsers that pay attention and can respond to DTDs before we worry about expanding the common pool of generally acceptable tags and attributes. As long as we allow browser manufacturers to decide which syntax to support, we'll be dragged through the tar pits of cross-browser performance issues.
As soon as we declare "unacceptable" those limiting programming decisions we will be opening the floodgates of true "mass publication", freeing those who have been trapped in the "my browser does this when the code does that" petty arguments for the past nearly 2 decades.
"Charter is notifying customers affected by its NebuAd test, while pointing them to a page where they can opt-out of the service."
Yes, but ... the service isn't running yet, and when one goes to said page and opts-out of said service, they are told to try again later when the service is running.
And it's the same type of "opt-out" mechanism as was proposed by Phorm, in that it's a cookie, so you need to "opt-out" from any browser you might happen to use on every system you use.
For those of you (@ZM) who comment about the absence of an outcry on this side of the pond, well, there really hasn't been any notice of this at all, over here, unless you happen to dig into the Charter website, which few, if any, do. I heard about it from the Spamassassin mailing list, for crying out loud! And I don't know what the British press is like on these types of hot topics, but this type of geeky stuff is pretty much restricted to geek news outlets, anyway. Frankly, I'm proud of how quickly our elected reps have jumped on this ... especially when compared to the speed at which your elected officials have (under)reacted to Phorm and its much wider rollout.
Paris, because she is an American who doesn't mind people spying on her habits ...
This isn't about PHP, but rather the exploitation of an application which happens to be written primarily in PHP, so let's not blame PHP, either. Turn the lasers on those sysadmins who have failed to protect their visitors by not patching their bulletin boards, or even, at the outside, blame PHPBB for the way they coded their product. It's not the first time PHPBB has been a conduit for exploits, and it's not likely to be the last.
Nice knee-jerk reaction, there. Not much for thoughtfulness, are ya?
Having a bootable stick or a LiveCD is extremely handy for things like, oh, I don't know, debugging your friend's Windows installation after it becomes unbootable, and bringing largish collaborative works into a workgroup environment, among other things.
Keep your HDD data to yourself by encrypting it like any reasonably-competent sysadmin would, and you'll be just fine ... your porn collection still private, and your virus all taken care of by your buddy with 'Nix-on-a-stick.
"So far, the countries have provided little proof that Chinese hacking is any different from cyber operations being conducted by other governments."
Such as ... ? Are you gonna back that up, or what?
It presumes that "other governments" are, then, attempting the same types of attacks as those the Chinese government is being accused of perpetrating. Where is the outrage, or at least a link? Tsk tsk.
The point is, why cart off the whole thing when there's no evidenciary value to it?
If on-scene forensics can determine whether the system is worth seizing or not, that works a lot better than seizing everything and then using lab tools to figure out what to keep and what to give back to the suspect. And, yes, the suspects generally get to have their equipment back if it is useless as evidence and if they ask for it, at least in the U.S.
According to researcher Ronald van den Heetkamp, as provided in the link in the article, the biggest risk is to sites running MSSQL servers due to the way they handle this particular query effort:
"...as you can see below the SQL query -or stored procedure- is almost fully HEX encoded, which means that no single quote is being used. Casting in SQL server or simply HEX() or CONCAT(CHAR(),CHAR()) in MySQL is widely known, and a good alternative when single quotes are not allowed upon injection, which makes it far more reliable. In the case of SQL server -which allows query stacking by separating the queries- this is crucial for a guaranteed compromise, and hence the reason why MSSQL sucks, and MySQL or Postgre rocks."
This would explain the majority of .ASP pages you see ... they're running the Microsoft server and Microsoft's database. Does anyone still have any doubt that Microsoft's server products are more prone to successful attacks than others? If you do have doubts, please explain why, and how you got to be in IT in the first place.
Not sure that it's a "user request" issue, rather more like a "overseer" issue.
Since RHL doesn't provide any "phone home" services in its default installation, the US Gov't was chaffing at the thought of providing low-cost systems to low-income people with no "anti-terrorism" mechanisms.
Since OLPC is a bunch of administrators (sorry, Nick), and I don't mean system admins, they don't have the ability to put such a mechanism in the PCs.
By installing Windows (I guess they've figured out how to unbundle Media Player?), the brass gets their spyhole pre-installed and quite robust without needing to develop anything or modify the base installation.
Microsoft extends their market into areas that previously had no business spending the required amount of money to join in, and the US Gov't gets their Windows on the World(TM) without needing to do any heavy lifting.
Oh ... and I agree with cam re: developing a Sugar-free alternative.
From Rick Rhino at another website ... a great idea for those of you considering trying to test whether encrypted files would cause a problem at the border:
"Take a family photo (or more likely multiple photographs) with a digital camera, and take their encrypted data... then take each bit of that encrypted data, and add 1 or zero to the red channel of that pixel, corresponding to the bit from the encrypted file.
The photographs with the encrypted data will be indistinguishable from a normal photograph, as that one bit is below the noise level of digital cameras.
You copy these photographs to your laptop and travel to the U.S.. Meanwhile, your contact in the U.S. has a version of these photographs without the encrypted file superimposed. Diff the files, and you have that encrypted file again.
You can do it with any sort of file (sound, images, etc.). Do it with a music file in your iPod, which is unlikely ever to be searched, and even if it is, they have only discovered a perfectly normal audio file."
Thanks, Rick!
Unless Microsoft engineers have some sort of magic pill, their equipment/OS will be as flaky as any of their customers. This means we can expect a data-destroying crash scenario every couple of years, or less.
Not to mention what would happen should their MS servers get hacked ... a not-unlikely proposition.
I'm not a fan of any project that stores user data online, beyond encrypted login data and some other smallish items. 5GB of personal and/or business data is a lot to risk. That goes for Google Apps and RIM, too.
I think the reason people prefer being scanned to being patted down is the simple fact that they get to avoid contact with security personnel. In addition, the images aren't detailed enough to illustrate which size condom you would need, and facial features are fuzzed in the newest version (not the backscatter version, which can't detail the face, anyway), so relative anonymity is assured. If a non-organic element is detected, the scannee is still within the scanner (not the backscatter one), so you're not going anywhere. Personally, I would prefer to be scanned rather than submit to an inefficient frisking by TSA hacks.
There's not much chance of avoiding this type of technology as time goes by, and these implementations seem to be more efficient and more successful than relying on humans to do the searching. I find that comforting.
The TSA has been around for less than a decade (since just after 9/11/2001), and, when left to their own devices, have proven to be somewhat less than professional. Read this disturbing collection of TSA gaffes for a few details: http://www.reason.com/news/show/29034.html
@1) Hunh? Nice retort! LOL!
@2) Are you familiar with the technology you mention? Probably not ... it's called "analog" technology, and perfectly accurate copies are not possible using it. In addition, spreading hundreds or thousands or copies around used to be quite expensive, what with the cost of the tape and the postage and all.
@3) Do you mean the music industry that's failing to stop your theft of their distributed product? How would you propose they pay the musicians, then?
Your arguments ignore the huge difference in the scope of the theft that used to fall under the "personal use" copyright guideline, and which, today, falls under the "unlawful distribution" section of that same set of laws.
It's not a matter of a single fan making a "mix tape" so they could play the music from that 12" vinyl disc in their car. It's a matter of some punk attempting to gain street cred by uploading a disc's worth of perfect digital copies onto the Internet, or downloading same, while puffing out their little chest and proclaiming "I can, so I do" and "Why should I pay? If the record companies put out better music, I would buy a disc, but since they don't, I'll just steal what I want."
Your arguments are specious, like every other freetard argument. Pretty selfish.
1) It's awfully amusing how many freetards include both "music industry is producing crap" and "I can download anything I want" in their rants ... so I guess you enjoy downloading crap?
2) Freetards are almost exclusively kids, because their friends and they "grew up" around others who also felt no moral compunction about the theft they so enjoy. Ask those of us who "grew up" before digital music, and you get a different impression about the acceptability of stealing an artist's work.
1/2) Artists of quality do not "work for a couple of days and get paid for a year", however much you'd like that to be the case. They have spent their entire lives developing their skills and channeling their creativity, only to have their work product, as valid as those hamburgers you freetards spend your days flipping, stolen out from under them.
Billy Bragg does not earn a lot of money ... perhaps he earns more than musicians you haven't yet and may never hear from, but he's hardly rolling in lucre. Every download not purchased impacts directly on his bottom line, and even more so on those artists who are trying to make enough to support themselves.
All you freetards deserve the future you are rooting for ...
Anyone who thinks stealing a copy of someone's art is just fine is, indeed, retarded.
Retarded in the sense that they are very slow to understand the reality of the situation: There IS no "original" when it comes to music, stupids! A television is a corporeal object that is able to be stolen by carrying it out of your house. A song is able to be stolen by making an unauthorized copy of it. Duh!
Anyone who downloads music or video or any number of other intangible art forms without the artist's permission or payment to the artist should be dragged naked through the streets and have a video of their "performance" posted on YouTube with the tag line: "Look, Ma! No brain! (and no pants, neither)" You are all morons and selfish bastards. May a cat sh*t in your boots.
Can't make your own planes.
Can barely make it through a week without losing a disc full of unencrypted data.
Can't get your criminal divisions to investigate clear violations of your own RIPA.
Let's just insult the US. It's a lot easier than being constructive.
Hmm. Hee hee. Have a nice day! Hope it's sunny for ya ...
Oh ... BTW, the USAF/Airbus deal went through BECAUSE of our equal opportunity capitalism, not in spite of it.
@Samuel Walker
Do you know of any examples of "true" random number generators in the programming world? Or even the distant hope of one?
@Brett
Good on ya! Credit where credit is due. Many articles about things like cache poisoning, spamming and other internet-age bugaboos fail to mention what is available to fight the beasties, even as those who are responsible for implementing those solutions (or at least the next generation of potential solutions) ignore them. DNSSEC for cache poisoning, SMTP-Auth for spam. etc.
When a potential solution will cost real money to implement, the big players avoid adopting it, choosing instead to hope nothing really bad happens or to push responsibility for fending off the problem onto some other player.
"Users, activate those phishing filters, now with cache poisoning detection technology!"
"Users, buy our latest anti-spam filter!"
"Anti-spam filtering services such as MessageLabs"
Our number one spam source for many months on end has been MessageLabs' server farms. How dare they criticize Gmail et al. when their own barn door is wide open!?!
In the absence of SMTP-Auth or restricting outbound mail in some other immediate fashion, there's nothing any of the aforementioned providers can do to stop spammers.
During his round of media appearances following his appearance at Congress, Bill spoke on National Public Radio about the IT worker issues.
During that interview, he argued that Microsoft seeks the most talented Computer Science graduates from the "best schools in the world", which he repeatedly explained are those in the U.S. .. his words. And the "problem" his company faces is that many of the most talented graduates are not native to America, and so are here on student visas, which expire when they graduate, forcing those top-level graduates to return to their own countries and await the granting of a work visa, like the H1-B, for which there is both an exceedingly long waiting period and an exceedingly limited number.
By the way ... there is an exceedingly limited number available because the U.S. government claims it wants American companies to do a lot of hiring of American workers, so if you're out of a job in IT, you're either living in the wrong place or without the proper credentials ... according to them.
Once Bill's company hires a top-level Computer Science graduate, with starting pay of at least $100k/year, from one of the "top schools in the world", he claims to build groups of additional employees around the treasured CS grads, to support them in their ground-breaking work. He argued that his company prefers to hire American talent in these roles, and that there are very few American top-level CS graduates available.
So, to paraphrase, "Give Microsoft more foreign worker visas and we will hire more Americans, too."
It is not enough for Microsoft to establish corporate presences in other countries ... getting all of those American support personnel visas to do their jobs in almost every other nation is far more of a hassle than getting the foreign nationals to work in America. Think about THAT for a moment as you bash U.S. visa policies, all of you Brits.
The "tech investment gap" referred to in this article's headline, in Bill's robust capitalistic mind, would have to do with both the investment in foreign workers' abilities to maintain their jobs in America and in American scholastic achievement. Congress can only help with the visa element. Excelling in the scholastic element is up to those who seek work in computing-related fields.
Perhaps MS means "approximator", rather than "translator". Have you worked with their "translator" for .docx files? Crap. And that's their attempt to "translate" their own format from Office 2007 format to an earlier Office version!
Fortunately for everyone, they should soon be abandoning their attempts at "standardizing" desktop applications in favor of pushing the bulk of their programs onto the web, where more-focused minds have greater sway with regard to "interoperability".
AT&T is one of Level 3's top 3 partners, so while they may not own those loops, they control them. Not to mention that they are Level 3's #1 wireless partner. You'll discover the same sorts of partnerships with every major backbone provider, here in the US. And Hibernia doesn't even count, unless you live in a very small sector of the country. I realize that there are other providers on the partnership same level (Comcast, etc.), however AT&T is by far the biggest, and that's not even counting the loops they were forced to divest which were picked up by Level 3 and others. Are you sure AT&T has no opportunity to apply this plan to those networks? I'm not.
Aside from that, I agree with those who are coming out against this type of packet spying. While the speed issue might come into play at some point, it is far more irritating to know that any packets are being analysed for these types of indicators.
In the USA, AT&T owns nearly all of the loops that make up our network. For example, if I buy T1 service from Megapath, they lease access from Verizon, who leases their loops from AT&T. So, it's not about creating a new ISP or finding a way to work around AT&T's cabling ... they have access to all of it because of their unique position in this country. Maybe even in your country.
This makes Jon Green's comment about losing "common carrier" status more relevant, because AT&T is classified as a common carrier, here. It just might force a showdown over the classification of the Internet itself ... private enterprise or public infrastructure? Paid for with taxes and usury fees, or by subscription? Required by fiat to be available to the public to freely use, or restricted by policies drafted by a single corporate entity?
This type of incident may be a grand opportunity to finally equate the Internet's wiring with the asphalt upon which we drive our cars. Where it has been the property and under the control of those who built it, now may be our chance to force its classification into the public realm, and force AT&T to give up any control it has over the system, making them publicly-funded stewards rather than privately-controlled overseers.
(It's a BSD kernel ...)
And system ... Posix AVs don't need to run as root, they only protect the userland. If anything gets past that, into the OS core layer, then it's not a virus. It's a rootkit or whathaveyou, and any modern distro comes with some hardening and antirootkit stuff. And, btw, my Linux boxes loaded with Clam AV and Panda haven't seen anything challenging in the past few years. Posix desktop users don't usually need to run servers (bind, etc.etc.) any more than Windows desktop users need to run Exchange Servers or IIS. Besides, Posix anti-attack progs are far superior to anything in Windowsland, because their programmers understand security better and are supporting much more secure systems out-of-the-box than Windows programmers possibly could.
Flame away!
Setting Kaspersky AV to delete anything it deems suspicious is an incredible show of faith in its accuracy. Setting it to quarantine suspect items is much safer, and explorer.exe could have been simply recovered using the recovery shell, could it not? For that matter, if one knew what had happened, simply extracting explorer.exe from the same recovery shell would have fixed things right up.
Probably the bigger issue was with not knowing what had happened, and being unable to contact Kaspersky to find out.
@Shakje
First, for your dictionary, it's a "buffer overflow". And it is one of the top attack vectors on any system.
Second, why isn't there a UAC on my Linux systems? I'll tell you why ... it's because the OS core layer is completely separate from the application layer, similar to the BSD clone used in OSX, and there is a strong, well developed security structure (multi-user separation, for one) in place. These factors are also contributors to the paucity of malware for Posix systems, as it is difficult and far more complicated to get a toehold into the OS core layer as a result.
This is very different from the Windows structure where many Microsoft applications are tied into the OS core layer, and where a "buffer overflow" at the Microsoft application layer can often easily compromise the OS core layer and allow for, among other dangerous things, privilege escalation, compromise of the Windows kernel, and on up to the compromise of even the BIOS and CPU.
If Apple is using similar hooks into their OS core layer, then Apple application security issues also present a significant risk.
While your statement about few users does apply to Apple users, if your explanation for fewer pieces of malware and its relation to "not as many users" were to hold up, you would need to explain why so few web servers (apart from Windows servers) are compromised in the OS core layer. Sure there are apps (notably forums, AJAX and sendmail) that show weaknesses and can be compromised to swamp the resources of any server, but the majority of web servers are running Posix (Unix, Linux, etc.)
Your argument would dictate that, because there are many more Posix servers out there, they should be riddled with malware, since they are a much juicier target than a home user's system. And yet, they are not. Because the systems are fundamentally different from Windows, the "come on down" king of malware.
And please don't call me naive. It's so ... belittling, and really not much of an argument for your points. I hope I have not been similarly disrespectful to anyone here.
Thank you, Mr. Faison (I do apologize if it's the wrong salutation.)
I understand your point of view.
I was just this past weekend discussing with my wife the amusing and perilous new 'security' features of Windows Vista, in particular the UAC, that substitute reasonably secure programming for a bunch of pop up windows prompting the user to 'ok' various procedures.
I believe our consensus was that the burden of providing security for any individual machine is being transferred to the user (i.e. "Hey, man, you said it was okay to install that spyware.") rather than being taken on by the providers of the software that makes the machine go. The software makers claim that they want people to use the computer as an appliance (what operating system does YOUR phone use?), but they are unwilling to accept responsibility for the yawning gap between what they sell and what the buyer must then commit to do in order to maintain that illusion of reliability.
Ah well. I guess that's what constitutes 'freedom' in today's computing environment. Let updates be updates, and let the Average Joe take some kind of online security class, if they give a darn about their role in the larger community. If they do not, we have only them to blame ... for clicking on one too many pop up windows. Or not, as may be the case with these and probably many other security updates.
And I don't want to start a flame war, but it is the knowledge of what one is commiting to as a computer user that often makes the Linux system more reliable than the other two big guys ... those who take the time to learn how to use a Linux system know that they need to continue to work on their systems to keep them clean and efficient, whereas Apple and Microsoft users tend to believe that their software overlords will keep the Huns from their gates, and inevitably pay the price for that belief. I use all 3 operating systems in various circumstances, and they all have certain benefits of use. And all 3 need user attention and education in order to perform at their best.
Buffer overflows affect the application layer, not the OS core layer. That is why (need I tell you?) Apple issued patches for both OSX and Windows. It is on the programmers of the applications to handle these types of flaws. As you point out, writing proper handlers should be part of Programming 101, by now.
As far as going head to head with a Turkish hacker (is there any other kind?) ... been there, done that. Just trying to keep it simple for those here who do not have a clue, as, obviously, many do not. I could write my own opus on this well-known application programming issue, but the Wikipedia article seemed to cover some good, basic ground ... again, for those of us who have not taken the time to wrap their heads around this serious and obvious security issue.
See my above post and the vast body of information on the rising tide of home users whose systems are fast becoming part of the zombie nation to determine whether this is a vapid claim.
Because, Mr. Davis, the biggest problems we face on the Internet today have to do with (a) spam/phishing/garbage emails and (b) identity theft.
The Average Joes mentioned by Mr. Goodin in his article, none of whom are reading these notes or even are aware of the nature of the network in which they play, let's their computer get taken over by exploits JUST SUCH AS THESE. Exploits that appear to be benign (in this particular case, through lack of any indication of the potential severity of the bug) yet which allow the most inexperienced script kiddy to add that Joe's computer to the zombie nation.
Then THOSE computers are used to (a) reach out and take over other systems, (b) send any personal data found on the now-hostage system to its botmaster and (c) start spewing a flood of crap that binds up the unsuspecting Joe's computer and the network at large.
This is a VERY SERIOUS (note the caps ... that means "serious") issue. Average Joes DO NOT know what the hell they are doing. Apple and Microsoft et al. are the CREATORS of these problems by their obscene refusal to do what is best for the common good, therefore it is incumbent on the crappy programmers at Apple, Microsoft et al. to do their part by HELPing the average Joes to keep their systems secure so the rest of us ("above average Joes") don't have to pay the price for their incompetence.
As Mr. Goodin complained (again), the pitiful "please install this, or not" notices available to anyone who believes that the alerts tell them what they need to know (Average Joes) were simply not alarming enough for this issue. They SHOULD have said something like "This is an EMERGENCY, CRITICAL download that you MUST install IMMEDIATELY in order to minimize the chances that your computer will be taken over by criminals, causing you to spend money getting your issue fixed, and potentially causing serious issues with your Internet community. DO NOT WAIT! Download and install this patch RIGHT NOW!" They could even restrict any other activity with the affected applications until the patch was applied, as demonstrated by their response to the QuickTime Flash player issue noted above.
It doesn't matter if it's Apple or Microsoft ... neither does a good job of informing its most-vulnerable users. And if Apple can't do any better than their ex-rival, now-partner, then they deserve to go straight down the road to Hell that has been so neatly paved by their compatriots at Microsoft.
You seem to be very good at ferreting out Mac support docs, and yet you still have questions about one of the most dangerous and prolific types of computer exploits. Spoon-feeding is no way to educate yourself ... you need to get more information about the issue, and not just sit there defending what little Apple has done to protect you.
As Mr. Goodin complains, the notification that the patch is available should STRESS that it is CRITICAL to install the update, which is does not do, even in the depths of its documentation, as reported by others in this thread. The screenshots of the notification alerts CLEARLY suggest that the patch is NOT critical, and can easily be postponed at the whim of the user, increasing the likelihood that such patches will be belatedly installed, if at all.
Apple is reporting the very most basic info about the bugs their security patches are intended to fix, namely that buffer overflow conditions cause the application to crash. What they don't tell you is that when the application is crashing, your computer may be in the process of open the gates wide for the barbarian hordes. Here's a basic description of a buffer overflow, for your edification:
"A buffer overflow is an anomalous condition where a process attempts to store data beyond the boundaries of a fixed-length buffer. The result is that the extra data overwrites adjacent memory locations. The overwritten data may include other buffers, variables and program flow data and may cause a process to crash or produce incorrect results. They can be triggered by inputs specifically designed to execute malicious code or to make the program operate in an unintended way. As such, buffer overflows cause many software vulnerabilities and form the basis of many exploits. Sufficient bounds checking by either the programmer, the compiler or the runtime can prevent buffer overflows."
And yet, Apple did not include "sufficient bounds checking" until, perhaps, these recent updates. (Agreeing with the sloppy programmer comment, above. This should be very old news, for them.)
See the illuminating yet hardly exhaustive article at Wikipedia (http://en.wikipedia.org/wiki/Buffer_overflow) for more low-level info on buffer overflows. There are many excellent articles and even entire books on this subject, and it is addressed in depth in any hacker guide for dummies. (I recommend the not-for-dummies "Stealing the Network" series of books ... by hackers, for IT security-types.)
And don't EVEN get me started on stack overflows!
You're misinterpreting the term "bundle".
Opera is not talking about whether a program is installed on the computer or not. Microsoft already includes both AOL and MSN on the desktop of most new systems (in the US), and THAT is what you all think "bundling" means in this context.
It ain't. (heehee ... could not resist).
Microsoft adds proprietary hooks into their proprietary subsystems that are unavailable to any other browser maker. Firefox cannot get at those same hooks, and neither can Opera. This puts them at a distinct disadvantage with regard to integrating their core programming with the operating system. It is an advantage that Microsoft has taken at every opportunity.
What Opera wants Microsoft to do is remove those hooks so everyone is on a level playing field, or allow some other browser maker access to those hooks so at least they can give Internet Explorer a run for its money. THAT is what "bundling" means in this context.
If Microsoft "unbundles" Internet Explorer and forces it to run on top of the OS, in the application layer like everyone else, then a lot of their custom web tricks will fail, and they will NEED to adhere to the standards in order to achieve the same functionality. If Microsoft "bundles" an alternative browser into the OS core layer, like their own browser is, then at least that one player will be able to compete.
Microsoft will ALWAYS have a browser "installed". The question is, is it taking advantage of their humongous, perhaps monopolistic market share to crush the competition by hiding and exploiting "trade secrets"? Are they willing to go head-to-head on a level playing field with another, core-layer browser?
My guess on both fronts is "no". They can't stand to reveal their hooks, and they couldn't win in a fair fight. Ever has it been so.
Oh ... and they proved they could make a (crappy) browser that IS "unbundled" from the OS core layer when they released Internet Explorer for the Macintosh. Now if they could only figure out how to make an OS without one, they'd be ready to play.
"Pure Java", if developed using the "correct" (non-MS) specs will truly run anywhere there is a "true" (non-MS) JVM.
And the biggest difference between Java and .NET is ... the net ... Java doesn't need it to run whereas .NET needs Windows and ... the net.
And Tim Anderson is right, following on the heels of the also-right James McGregor. As for AJAX performance ... ye reaps what ye sows.
So ... don't add that particular Google-ian anchor attribute if you don't want to. No big deal. Your pages will validate without it, and the Big G will treat your site only a little differently from those who want to please Her. You'll still have Yahoo and MSN.
For those who don't yet know, the attribute Google wants us to add to any of our "paid links" (including unpaid-yet-potentially-beneficial-to-the-receiver outgoing links) is this: rel="nofollow", as in <a href="http://notmysite.com" rel="nofollow"...
This simple (and, granted, out-of-the-spec) use of the rel attribute won't hurt you in any way except that you'll get those links flagged in a strict HTML validator. Other search engines won't mind, and Google will reward you by not following those links and thanking you for your effort. Screen readers don't give a darn, and I can't find any application that objects to this, any more than they object to the hundreds of MS-specific tags and attributes that currently clutter up the web.
Google simply wants you to tell them not to associate that link with your page content. "Google: Don't follow this link with your spider." Simple. Don't use it, if you don't want to. Your paid links will still get the exposure and visitor clicks. Done deal.
And it's hardly MS-ian, as their mantra is more like, "Our way or screw you."
It's almost always tough, when a new player starts making headway.
1) General Motor's EV1 Electric Vehicle
They made under 50,000 of them, got a bunch of metropolitan transit agencies to pony up for the charging systems in public areas like at airports and so on, let the people who leased them have a great time with them for a couple of years, then recalled all of the cars for crushing (!) because "there wasn't a market for them". Despite the fact that literally every EV1 user loved it and that there were outstanding orders for more than 500,000 more.
2) Honda's new Hydrogen Vehicles
Has anybody seen the new hydrogen refueling stations? I haven't, either. I wonder how long these will last?
3) Dell + Ubuntu
As noted above, there are issues with locating and purchasing these machines, built on low-end hardware, resulting in relatively low numbers when compared with MS-installed product lines. Many people complain about how "hard" it is to install 'Nix and how their pre-installed MS systems work great. They have no idea how well a pre-installed version of 'Nix would be, because the distributors don't work at it very hard. Dell could quite easily put together a gaming MONSTER machine with kicking graphics and tons of power (plus WINE for PC gamers, wifi, bluetooth, et al.), and then simply tweak any 'Nix distro they wanted to until it worked perfectly on the machine ... then image that installation, build a bunch more killer machines and clone the image onto them. Doesn't get more simple than that, and it's exactly what they do with a Windows offering.
The new guys have to get the corporations to take more of a chance with them, because it's a tough business decision to spend a ton of money promoting something outside of the public consciousness. Shareholders don't like risk.
On the good side, however, I don't recall any major vendor even considering Linux pre-installed before 2005, so things are moving in the right direction. Maybe by 2010 we'll have a decent selection from which to choose.
(PS: Me = IT Director/250 XP+Vista workstations/8 Fedora servers; Home = Fedora/BigSlack/XP Pro/Win98 multiboots ... whatever works.)
@Sabahattin Gucukoglu
DSCP needs more implementation to become truly beneficial across all subnets, and it is probably the primary mechanism through which ISPs will "rate" and then charge their clients for, in a non-neutral world. It's a great idea, though ... even if Windows 2000 does tag all outgoing packets as Class 1 (IP Precedence 5) content regardless of its true nature.
@Morely Dotes
You're right ... fatter pipes would do the trick.
Speaking of not understanding how the 'net works ... BitTorrent cannot "saturate the pipe" except for in one segment: Your local CABLE loop. If you've got a pile of kids downloading torrents in your neighborhood, that has absolutely no impact on anyone in another loop.
And this ONLY applies to a CABLE connection, which is and has always been a shared resource. You would do better to blame your cable company for overselling your local loop than to disparage the robust nature of the network, itself. That shared resource gets slower with every additional subscriber on your loop, and even slower when those subscribers get the TV/phone/Internet packages that are all the rage, these days. And those subscribers who are grabbing the HDTV feed? They are requesting more data than any torrent user.
Try a non-shared resource like DSL or T1 or something if you're tired of the pirates on your block slowing your own data grabbing.
Net neutrality is an important battle for every web publisher. Unless you've got the deep pockets of a major corporation, your voice will be crushed under the weight of onerous regulation, and your content will be analysed with each request. Say 'bye-bye' to independent thought!
From the Windows Server Division blog, linked in the article:
"In fact, this very web site ... has been running entirely on Windows Server 2008 since before Beta 3."
Now THAT's an endorsement! It shows that the latest version of the Windows Server line is FULLY CAPABLE of hosting a blog, dammit. In it's entirety!
Take that, MS doubters.
(It's also fairly amusing that despite the blog's statement that one of its purposes is to "to provide a place to have an open and honest dialogue about the development process", ALL of the comments are trackbacks to other MS blogs ...)
And my wife makes an excellent brisket by braising and simmering. No baking required!
Windows users please note that when your system reboots to apply one of the popular "security patches" ... that restarts your 'uptime' clock. So unless you haven't been updating your Windows installation, your 'uptime' figures are probably no longer than 60-90 days.
Linux only needs to reboot when you update the kernel ... My Fedora servers have uninterrupted 'uptime' of over 3 years. No runs, no drips, no errors.
Also, sit a new user down in front of any operating system and leave them alone, and they'll probably have a hard time getting going. Computing is still nowhere near transparent enough on ANY platform to be considered 'better' or 'worse' than any other platform.
When was the last time you thought about the operating system used by your land-line phone? Talk about thin clients ... When day-to-day computing reaches that level of transparency, threads like this will be achingly dull. And if the operating system in primary use at that time is a Microsoft product, I'll eat my phone.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
