Trusted news and trusted reviews

Readers can't pay, won't pay for content

Mike

You wrote:

"It doesn't as a CNN executive was quoted as saying this year, have to come from a "trusted brand name" like itself or Fox News..."

Wrong. The Inquirer and The Register are trusted brand names, as far as I am concerned. CNN is dodgy, and Fox News is up there with Pravda and Izvestia. ("There's no 'fox' in Fox News..."?)

Other trusted brand names (of mine) include Brian Cloughley and Theodore Dalrymple. The great thing about the Web is that it lets you find the really good stuff that's out there... and gradually, by word of mouth, people gravitate to those sources.

Tom

I've been in the IT business for 30 years and nothing much has changed in that time except that now everyone with a glorified typewriter thinks they know all about computing -- all of computing that is: if it works on a PC why can't it apply to a corporate network. But that's another rant...

I remember when Computer Weekly and Computing were on free distribution: why would anyone want to pay just to read job ads? The rest was of no consequence, but help to distract attention from all those blinking lights on the night shift. A few years ago, after some absence from the job hunt scene I discovered they had become titles that had to be paid for (even wrapped up in the newsagent to discourage browsing); strangely this coincided with the disappearance of much of the recruitment advertising so they didn't get my money. The last time I looked at a copy in the staff room of my university's CS dept there were no job ads at all -- so now I don't even pick them up. Why should I when I can get all the gossip etc from the comfort of my desk whilst appearing to be busy at work. (night shift redux)

I don't know how you'll make a living. I have subscribed to Linux Weekly News. Maybe I would subscribe to the Inq and the Reg, but it needs to be painless; only a little scratch on the plastic, no complex logins, (and don't overlook I use Opera and delete all cookies on exit). And that still leaves the problem of content: I want more than an aggregation of press releases, giving them a cynical tweak is not enough. People pay for information and advertising can be informative and even useful if it dosn't flash in y' face.

These are hard times in the IT world. That covers not only more than half my lifetime but that of EDP itself. I do the sort of sys-admin work I went back to university research to avoid, and earn less than I did a dozen years ago. Ah, well, best of luck to us all.

David Clark. Systems & Web Development Manager

Your one brief mention of micropay is the core of the problem. We don't have any way (yet) to make micropayments work. And that's essential to providing support to the numerous small outfits where we all browse. Solving micropayments isn't a trivial problem. Keeping costs down while handling millions of small transactions securely, with an audit trail, subject to customer review is going to be a big challenge. The person or company who makes this work is going to become very well off. Suppose Magee charged $.01 for each day a person browsed TheInquirer, all you want for one cent. In a year it would cost me about $3.50. That's not a bad cost, considering. Today, you could make that work with a charge or paypal for a 1 year subscription at about $4.00, to cover processing. That's still okay in my book, but probably not for occasional visitors. They need micropay. We all need micropay. The side effects on advertising would be memorable. The shake out of the good, bad and ugly sites would change the landscape, maybe for the better, maybe not. A lot of us would eat better, some not so well. All in all, it is needed. After all, consider micropayments as a micro version of what you do in the 'real' economy. There's gotta be a way. Bill Nicholls http://www.billswrite.com PS. You might be interested in my take on our economy, hurt from outsourcing. Part I: http://www.billswrite.com/bwrite/economy03a.html Part II: http://www.billswrite.com/bwrite/economy03b.html

IT salaries in India and USA contrasted

Just so that it was said, in india someone who gets 8,500 USD a year is rich. AKA they actually have a house and can go to the market in nice clothes to buy their food, and they still had to go to those mass-producing programmer colleges. They are not individuals.

A man in the US getting $8,500 is probably working 2 or 3 other jobs, probbly didn't finish high school, and is lucky if he doesn't live in a trailer.

If this doesn't stop here, there will be some sort of mass march against these kinds of employment, probabally resulting in a 1930s like hatred of getting involved with the world outside our little continent. If only the companies would realize the folly of their ways, and giving American jobs to foreigners (in these kind of numbers) is just wrong?

I'm guessing this will catch on quickly, but why does it matter when we will all be replaced by robots in a few years (http://www.marshallbrain.com/robotic-freedom.htm) ?

Very Concerned,
Caleb Fischer

Hi All,

This seems to be the in thing at the moment for many comps (Tiscali's call centre move, HP/Compaq...). If any of you hacks want the real story then you should look up Kevin Bales's book "Disposable People". The chapter on India is revealing. You might have a bit more to say about these multinationals, saving money at anyone's cost...or maybe not?

Regards
daily reader

Those figures just don't work. As more business moves to India the cost of living and average salary will rocket up. It will level off long before 2015 in the specific fields that are being outsourced. Right now it is steep because companies are willing to take a hit in one area to save money and keep them profitable so their stock prices don't plummet. As the economy recovers, they will find that they will start loosing market share because of the lower quality IT support they have and will find it harder to justify the savings. The same thing goes with China.

With the large population that is about 50 years or more behind the technology curve and just enough infrastructures to lure large businesses into the area to take advantage of the low labor costs will bring more money and technology into the area. If this influx is strong enough, they will have needs of their own to deal with the boom in the economy and technology that has just jumped up about 50 years. They will now have a large shortage of educated people to fill these rolls and the cost to employ these same people that outside companies are fighting for will drastically increase. Right now that is not the case, but they are assuming the salary difference between the two countries in those perfusions will remain as large as it is right now.

Todd Abbott

AMD64 is coming to a pc near you

It is very important that Microsoft ports Windows to AMD64. Without this OS port, AMD would only be able to sell to Linux installations. Microsoft isn't just doing this as a favor to AMD, they must compete with Linux and the performance gains AMD64 offer are too great to ignore.

Some important components of Windows being ported to AMD64 are the .NET runtime and IIS 6.0. While most web applications running on IIS today are ASP, very soon the dominant platform will be ASP.NET. Due to the great increase in capabilities, performance, and productivity of ASP.NET over ASP, and the stated direction of Microsoft, ASP.NET will become the defacto web development platform for IIS servers. During the transition to ASP.NET many of these websites will continue to rely on x86 COM components. Over time some of these COM components will be re-written in .NET while others will not due to performance needs or .NET limitations.

Pure .NET applications will run on the Windows 64 platform without modifications. (64-bit .NET) These applications will be able to take advantage of the AMD64 platform because the .NET runtime will be AMD64 specific. This can also be said for a version of Windows 64 for the Itanium. These applications will continue to rely on x86 COM components that corporations have heavily invested in. On the AMD64 platform, the COM components can run in 32 bit x86 mode while the pure .NET portions can take advantage of AMD64. This is not the case with Itanium as there is no 32 bit x86 support.

Now lets look at the companies with the most websites that are interested in improved performance. We are talking about hosting companies. These companies build an infrastructure of web servers and internet connectivity and rent these capabilities on a monthly basis to subscribers. In deciding on their next generation of Web hosting servers, performance is almost everything. A single server often hosts hundreds of websites. Increasing the number of websites that can be hosted on a server directly increases revenue. The AMD64 platform will certainly do this with it's increased performance and large addressable memory space. AMD64 and not Itanium will be the preferred platform for web hosting companies running IIS. On IIS web servers and any other pure .NET application, hardware independence will finally be recognized. As soon as Windows XP 64 is released, these hosting companies can take full advantage of the AMD64 platform.

Now let's look at another common server such as database servers. Database servers increasingly act as the backbone of many applications within a company. This dependence means the database servers must be increasingly reliable and quick. These types of applications are probably not good candidates to be written entirely in .NET and instead need to be written specific to the hardware platform to eek out all possible performance gains. In fact many database servers written in 64 bits have already demonstrated the benefits of 64 bit platforms. To compete against each other, database servers must be able to take advantage of the latest hardware. At this point, people will not need to rely on synthetic benchmarks to compare performance between microprocessors, but instead the database server software itself will be benchmarked and compared on different processors. Initial data suggests porting to AMD64 is much easier than IA64 and that AMD64 performs better. (IBM ports DB2 to AMD64)

Certainly most applications such as Microsoft Office and hundreds of other business applications have no need to be ported to another hardware specific architecture such as IA64 (Itanium) or AMD64. To allow these applications to instantly take advantage of new hardware, they just need to be pure .NET applications. This is the direction most developers will choose as .NET is replacing the older DNA model of software development. (Office.Net) As soon as the Windows OS and it's .NET runtime are ported to a new hardware platform, all these applications will be able to take advantage of the speed improvements. Incidentally this is also true for Java applications as they are also hardware independent. As far as benchmarks, the applications can be directly compared to each other on each platform. If any synthetic benchmarks are needed they will be .NET based benchmarks comparing common .NET tasks.

.NET represents the future of computing in that software will increasingly be written in hardware independent code such as .NET or Java. The main question on business developers minds is not whether they should code and compile for IA64 or AMD64, but rather how quickly they can re-write their code for .NET. Only the systems level programmers creating IIS, .NET runtime, or other special server software such as SQL server really have a need to code and compile specific to the hardware. While the transition to .NET based applications will take time, the AMD64 platform has the advantage of x86 backward compatibility. Once this transition is mostly complete, this advantage will be lost and all new microprocessors will complete on a more level playing field. This is ... can they convince Microsoft to port Windows to their hardware platform and achieve strong improvements on performance with direct application benchmarks.

In a pure 32 bit world where all your applications are last year's versions, you can still take advantage of performance gains of the AMD64. The 32 bit backward compatibility runs faster than AMD Athlons running at higher frequencies.

AMD64 already is demonstrating its appeal for massively parallel computers, 64 bit Apache web servers, 4/8 way multiprocessing systems, and 64 bit gaming systems. (AMD press releases) . Once Windows XP 64 is released, AMD64 processors will be well positioned to dominate not only over Itanium but also over 32 bit Pentiums. Whether your talking about servers, workstations, or pc's, the AMD64 is coming soon.

Ray Stratton

Hi

I think you seriously missed the point in your article "Microsoft Windows for AMD64 is just a first step". There's no need whatsoever for a whole slew of 64-bit applications to make Athlon64 and/or Opteron attractive to the Windows crowd once 64-bit Windows appears there. And suggesting that things like MS Office need to be ported is just silly.

The existing 32-bit Office software should run just fine under 64-bit Windows (if it doesn't, Microsoft's claim to 32-bit backward compatibility will look rather dubious). Same for virtually all the other existing 32-bit Windows applications that won't benefit in the least from being ported to 64 bits (at least in the Hammer environment: Itanic may be a different story, unless its 32-bit emulation software runs a hell of a lot faster than its 32-bit hardware emulation does): while it's certainly true that there'll be minimal incentive to port such applications, it's equally true that failure to port them will in no way detract from the desirability of the AMD64 hardware platforms - because they'll run those 32-bit applications every bit as well as anyone else (and better than some).

In all likelihood, most of the Windows software run on AMD64 for years to come will be 32-bit, even in 64-bit system environments. What will make Hammers more attractive than Pentium/Xeons (aside from any 32-bit performance advantage they may have) will be a *few* important applications that benefit significantly from a 64-bit environment (plus some more relatively unimportant ones - e.g., some games - that do as well). These are the applications that *will* be ported relatively soon (because the benefits of porting will justify the effort) - and any given customer need be interested in only *one* such application to make AMD64 a more attractive choice than Pentium/Xeon, as long as the prices are comparable.

What are examples? The operating system itself: Win64 can handle large memories (anything over 4 GB, and that's cheap enough today to start being interesting) better than 32-bit Windows can - both for keeping greater numbers of applications (even if they're all just 32-bit) resident, and for handling the system cache (which in 32-bit Windows is limited to using less than 1 GB of system virtual memory, and this can translate to far less than 1 GB of physical memory due to the granularity of allocation). 64-bit server (e.g., database) processes, even if the applications they serve mostly remain 32-bit. Photoshop is often mentioned as a candidate, video editing certainly qualifies, and at least some game developers have expressed significant interest (my background is in systems rather than in applications, so there are likely more pontential candidates than I can list off the top of my head). And, again, it only takes interest in *one* such application to make any given customer start looking hard at AMD64 - as long as there's no counter-balancing potential downside that they need to worry about.

So the list of applications that *need* to be ported to AMD64 is anything but 'daunting'. There should be no need to 'hybridize' applications: the same 32-bit application should install - and run - as well on 64-bit XP (at least on AMD64) as on 32-bit XP.

And 64-bit processing isn't primarily about benchmarks: if it were, SPARC would have disappeared long ago in the wake of IA32's across-the-board superiority in that area. Rather, it's about the advantage of a 64-bit environment for *specific* applications (that, indeed, may run much faster there, and/or may be far easier to develop there). So while AMD64's 64-bit benchmarks may often be no better than its 32-bit benchmarks (if for no other reason than that the latter can use Intel's benchmark-optimized 32-bit compiler, and Intel is not likely to be offering an x86-64 version any time soon unless Yamhill must be called up to do battle with AMD64), that just doesn't matter: what matters is whether some important applications can leverage AMD's 64-bit environment in ways that aren't available to them on IA32.

Several of those applications are already up and running on AMD64, just waiting for a production version of 64-bit Windows to appear. When it does, there won't be any need to wait longer to get a pretty good initial idea of what kind of customer acceptance the platform will enjoy.

Bill Todd

New vibrating power source discovered
Pyramids

Dear all,

Ref your article on a new 'vibrating power source' from Hitachi: just to prove there's nothing new under the sun, here's a couple of links to a very interesting book written by an extremely competent engineer who claims the Great Pyramid in Egypt was a monster, giant-sized one of these that powered the whole of Egyptian civilisation.

http://www.amazon.co.uk/exec/obidos/ASIN/1879181509/qid%3D1062325692/202-1779479-8150231

http://www.gizapower.com

(And no, the author's not mad or demented - the book is a genuinely clever and well argued piece that puts forward a theory that this giant edifice had a much more important and pragmatic purpose than a tomb and is supported by some rather compelling evidence)!!

If true, it would certainly prove the Japanese haven't lost their touch when it comes to miniaturising things!!!

Ian Wallwork

Sphinx

I did a Google search on "leaf spring" and found a page that described how leaf springs fail and it sounds to me like they should be part of a redundant system, so that when (not if) they fail, the results won't be disastrous.

Michaela Stephens

Government.exe corrupted. Reboot Congress? Y/N

I read on one of the forums or websites that XP SP2 may not come out for another year! Doesn't that mean ALL persons who buy their software are buying that vulnerability for another year, not to mention missing all the other fixes! Wow, new computer owners to will get infected needlessly! Maybe they ought to put on their ad pages "Warning: the Computing General has determined that use of this software is dangerous to your computing experience."

Somehow that seems to me to be borderline irresponsible, at the least, for a company that has so many relatively inexperienced users, you know, John Q. Public. By continuing to sell the same disc to as many people as possible, they're may make it easy for themselves to track fewer software versions, but an endless nightmare for their customers. As I wrote:

"If only M$ would learn from AOL and Earthlink and others who give away [free] CDs by the boxful (millions, I'm sure) at all the computer stores, and in the mail too. If only M$ would make a [free] CD that would have - on one CD - all the updates and hot fixes for Windows 98, 98se, Me, NT, 2000, XP, 2003, SqlServer, IIS, etc. What a publicity coup that would be!"

TekTips

By the way, notice I didn't get any stars. {sigh} The recipient of the stars did a fine job pointing readers to the instructions for slip-streaming, but my point was, what about the "uneducated masses" who don't deserve to be snubbed?

Mark

Wendy,

As always, I thoroughly enjoyed today's article in the Inq. I've been using email since 1982 (yep, an "old fart" in techo terms, despite my tender 39 years :-), and have avoided getting a single virus in large part because I have steadfastly avoided Outlook. That and I view all attachments as suspect.

Not sure what you use for an email client, but I love my Mulberry . I'm a happy convert from Netscape that I used for years until the dreaded AOL got hold of it. Basically I feel an email client needs to be pretty "dumb", and certainly doesn't need to be scriptable.

Your explanation of Behaviour Blocking was excellent, and would certainly seem to be the way to go for operating systems and/or the applications code itself. Actually, in thinking about it, I believe you'd want to have protection in both the OS and app. Outlook is in a much better position than the OS to know when its address book is being harvested and email is being sent in some automated way.

BTW: I'm part of a San Francisco startup (beprivate.com) that's produced an email transport subsystem that allows a user to set up policies that block certain email behaviours such as spam and harassing email. We also monitor and control certain valid SMTP protocol sequences that we see as behaviours typical of spammers. Unlike SPEWS we don't block email, we just become more and more uncooperative :-)

We're bootstrapping by offering an email hosting service using the technology. We've seen our system drop up to 70% of the email destined for some of our users, and all without compromising the privacy of the message itself (i.e. we don't read/scan/compare the message body).

Anyway, please feel free to look into it if the mood strikes you - though no doubt you're a busy person.

I look forward to your next article!

Paul

PS: Thanks to you, I'm now seriously addicted to Bookworm!!! My only revenge has been to get others hooked :-)

The best single change Microsoft could make would be to make it impossible for emails to do anything automatic, period. This includes not going out and getting images off the web. True, there will always be idiots that will click an attachment, but I don't do that and this one change would give me all the email protection I need.

Regards,

Greg

p.s. I enjoy your articles! :)

It appears that every time MicroSoft thinks about automatic updates as an opt-out mechanism rather than an opt-in mechanism, they come under fire, with words like "subscription" and "law suit" (okay, two words) being bandied about.

Wendy is correct when she writes: "Lots of us have little rules about when and how to apply updates."

As she points out, however, that is not the majority of users. To quote Wendy again: ""People don't know about it," said Simon."

If you have a rule about when to apply updates, opt out. Click the "ask me before applying any updates" button. Corporates have it even easier, as they do customized roll-outs anyway, and have admins that like nothing better than to play with their group policies all day (apologies to admins that don't).

It seems strange to me that MicroSoft offering automatic updates by default, with an opt-out mechanism, is seen is diabolical, while other companies doing so is hailed as innovation and user-friendlyness that makes those other companies' products stand a foot above MicroSoft's products.

To give concrete examples: LindowsOS reviews seldom fail to mention how wonderful the (mandatory, not opt-out) CnR auto-update is that runs at every startup, keeping the user's computer free of security holes without a need for user intervention. Mac users rave about the way their machines auto-update and keep themselves clean and trouble-free, "bwahaha, you poor Windoze suckers, muhaha."

But when MicroSoft mulls the idea of making patch management an optional rather than a mandatory chore, they get lambasted for it.

I truly, truly do not get that.

Regards

Shawn Behrens
Integralis/Activis Managed Security Services