WINDING DOWN
An idiosyncratic look at, and comment on, the week's net and technology news
by Alan Lenton
A somewhat belated happy new year to everyone :)
We're back with a round up of some of the more interesting things that happened in the last week. I did consider covering the whole gamut of things that happened over the Xmas period, but decided against it - this newsletter is long enough as it is, without adding the equivalent of another three issues to it!
It's been raining here, of course, being the UK, although at the moment there is a five minute break before it starts again, so I will take the opportunity to offer you this week's collection of soggy ones and zeros!
Shorts:
I'm happy to start the new year with news of an outbreak of common sense on a very contentious issue.
The Berkman Centre for Internet & Society was tasked with looking into the issue of the online safety of children. This all came out of an agreement between MySpace and all the state Attorneys General about a year ago.
The Berkman centre trawled diligently trawled through all the research available and came up with the conclusion that the real threat to children on the internet is - other children!
It seems that contrary to beliefs of politicians and judges, the internet isn't, actually, awash with predatory pedophiles, and that the risks faced by children on this front are no different from those they face offline.
Parents whose children use the internet need to be aware of the findings of this paper, because otherwise they may end up trying to protect their children from the wrong things, and therefore missing real dangers.
The issues discussed are too complex to cover in a short piece like this, but ArsTechnica has a very calm and considered two page analysis of the paper. It's well worth reading, even if you don't have children.
http://arstechnica.com/news.ars/post/20090114-report-biggest-online-threat-to-kids-
is-other-kids.html
And next, a piece about our old friends Microsoft. They've just released an interesting piece of software into public beta test. (Less of the sarky comments from the back about all Microsoft releases being public beta tests, please.) This particular program is for desktop virtualization, and it lets you run older programs under Vista and its successors.
Some time around the turn of the century a battle took place inside Microsoft between those who thought the new versions of Windows should be totally compatible with older versions and those who thought the old baggage should be jettisoned.
The compatiblistas lost and ever since then there has been no guarantee that even programs written to Microsoft's published interface specifications will run on future versions of their operating system. Now with the desktop virtualization program you can, at least in theory, run your old Windows programs. If it all works properly, it's a nice solution. I'll be interested to see what the results are like.
http://ifwnewsletters.newsletters.infoworld.com/t/4222725/250590949/163107/0/
The UK's Royal Navy celebrated the new year with a computer virus infection which shut down the administration networks on a number of its warships. The Ministry of Defence (MoD)* has informed us that there is nothing to worry about and that only 'a small number' of systems have had to have been shut down.
One has to wonder exactly how they would know, since they don't appear to have noticed the fall of communism some 20 years ago, and are still buying equipment to defend us against the Russian communist threat.
In the meantime, they've rolled out that hoary old chestnut 'security reasons' to cover up the extent of the screw up. The older I get, the more I become convinced that nothing changes (except the names).
http://www.theregister.co.uk/2009/01/15/royal_navy_email_virus_outage/
InfoWorld has a very amusing piece about the seven worst tech predictions of all times. Of course this only covers those that we have records of. Among the missing predictions are probably statements like 'This new fangled writing will never catch on' - unknown Greek bard. Or perhaps 'There is only a need for a world total of four of these' from a neolithic caveman gazing at the stone wheel someone just invented.
In the meantime take a look at the boo-boos we do know about, and shake your head in amazement. My friend Lois comments, '...I'm holding onto hope that the nuclear-powered vacuum cleaners may still be just over the horizon...'
http://www.infoworld.com/article/09/01/05/The_7_Worst_Tech_Predictions_of_
All_Time_1.html
There was an interesting, and probably very important, legal ruling came out of Florida when an appeals court upheld the decision of a lower court on the use of breathalyzers. The court ruled that for evidence from the Intoxilyzer 5000 breathalyzer to be used in more than 100 drunk driving cases, experts for the defendants must be allowed to examine the machine's source code for flaws.
The company involved, Kentucky-based CMI Inc, is refusing to obey a court order to provide defense experts with the code, and has so far accrued more than US$2 million in fines as a result. The company, and the prosecutors are using the argument that the code cannot be demanded, because it is a 'trade secret'.
I think Manatee County Judge Doug Henderson's statement on the issue, sums it up brilliantly. He said, 'The defendant's right to a fair trial outweighed the manufacturer's claim of a trade secret.' Exactly!
http://www.bradenton.com/847/story/1152077.html
I see the up and coming countries of Asia are truly learning from the example of western countries. As I write the chairman of India's top outsourcing firm has been banged up in the nick while awaiting trial for declaring a fraudulent one billion dollars worth of profits over the last few years.
Sounds kinda familiar, doesn't it.
I don't know how many well known public companies have outsourced their IT to Satyam, the firm involved, but I'd guess it's a fair number, given that Satyam was the largest of the Indian outsourcing companies. The problem now for those companies is how they rebuild their own IT systems in the current financial climate.
In some cases, where the outsourcing has been particularly ruthless, the real question is where to start sorting out the mess. These companies got rid of virtually all of their experienced IT staff as part of the outsourcing, and probably don't even have anyone left who is capable of recruiting new staff!
I'm sure the staff they got rid of when the outsourcing took place will be only too happy to come back - in return for cast iron contracts of employment for at least five years, twice the salary, and 35 days a year in holidays...
http://www.physorg.com/news150782544.html
http://ifwnewsletters.newsletters.infoworld.com/t/4209714/250590949/162207/0/
I see that a new York Supreme Court judge has given the green light for the state's Amazon tax. Amazon.com and Overstock.com had asked the judge to overrule the law which requires the companies to collect sales tax if they have affiliate networks in New York state.
At the moment residents who buy goods online are supposed to declare them and pay the sales tax themselves. Few, if any, do so, and the state governments have for some years been trying, without success, to force online suppliers to collect the tax.
Until now they have only been able to enforce collection if the online retailer has a physical presence in the state. The NY 'amazon law' extends that definition to include the physical presence of affiliates as being the same as a physical presence by the online retailer.
This ruling will drive up costs for both the retailer and the customer, so, especially given that it involves interstate commerce, I expect to see further appeals in the federal courts in the not too distant future. Watch this space for more information when that happens.
http://www.theregister.co.uk/2009/01/12/amazon_tax_ruling/
Not so much a rage against the machine as a rant against the spreadsheet from columnist John Dvorak. This year marks the thirtieth anniversary of the now ubiquitous spreadsheet program. According to Dvorak society has been going down the pan ever since this perfidious program unleashed a plague of accountants in the higher echelons of business.
A bit over the top, you may think, but my sympathies lie with Mr Dvorak. Time after time I find people swearing by spreadsheets that bear no resemblance to reality. Not only is the initial data false, but the assumptions coded in are simply wishful thinking.
Remember the old programmer adage, garbage in equals garbage out.
Take a look - it's a great little piece, and it doesn't use a spreadsheet (or a presentation) to make its point.
http://www.pcmag.com/article2/0,2817,2338796,00.asp
Homework:
Thinking of switching from Windows to a Mac in the near future? then you might like to take a look at an article on ZDNet by Raf Needleman, detailing his experiences in trying to make the switch. Raf isn't hostile - far from it, he went out and bought a pair of MacBooks for himself and his wife. However, things turned out to be not so easy as expected.
I can sympathise. When I was on a security course at Oxford University the class was using large screen desktop Macs, and I found the user interface a nightmare. For the record I regularly use three different interfaces (Windows XP, Linux/Gnome, and Linux/KDE).
Particularly unusable was the placing of the menu at the top of the screen regardless of where the window you were using was. I ended up with the menu in the top left of the screen being one and half feet away from the window I was using in the bottom right hand corner. This is a throwback to the days of 8-bit machines when you could only have one full screen application running at a time. Very primitive.
The article is very comprehensive, and as the author points out at the end, a lot of the problems are a result of needing to run specific applications. If you never used a computer before (i.e. you live on Mars) then you are not likely to have so many problems.
If you are thinking of making a change to the Mac, read this first and then go and read something by a Mac supporter, so that you get both sides of the story, before making up your mind.
http://newsletters.zdnetuk.cneteu.net/t/447147/921984/636070/0/
The stuff I look through for this newsletter is generally an example of incremental improvement, but once in a while something crops up which is a genius piece of lateral thinking.
You've probably heard of 3D reprographics (the 3D equivalent of a photocopies). What happens is that you take the data from your 3D modelling package, feed it into a 3D copier, and it builds you up a physical 3D version layer by layer.
Neat, but nothing spectacular, 3D modelling has been around for quite a long time, but only recently have the prices dropped to levels where it is starting to be more common. Now, in what I can only describe as a stroke of genius, a techie at pharmaceuticals company Glaxo, Smith, Klein realised that the data from 3D modelling packages is similar than that from CAT scans. So, he wrote a program to allow the modeller to build copies from CAT scans.
Brilliant! Because now surgeons who are about to embark on critical surgery can get a 3D model of the organ to be operated on and plan his, or her, strategy in advance.
[Source: A GSK contractor]
The programmers amongst you out there may like to take a gander at the SANS Institute's web site, where a list of the top 25 'most dangerous' programming errors has been published.
They include all the usual suspects - broken input validation, SQL injection, cross site scripting, buffer overflows, dodgy (aka home brew) cryptographic algorithms, and - get this - passwords hard coded into the software's authentication module!
While a lot of problems, especially those classed as 'Insecure Interaction Between Components', relate to the new generation of interactive web sites, there's plenty of mistakes for 'real' programmers to avoid. Among other I spotted improper initialisation and client-side enforcement of server-side security. I've definitely seen those before.
This list is good and will help enforce good programming practice, but there is the danger that it will become a set of tick boxes which people can use to cover their backsides without really understanding what the problems are. There is also the problem that it only covers dangerous coding practices, not the equally important dodgy program design practices. But it's a start.
http://www.sans.org/top25errors/
http://www.informationweek.com/blog/main/archives/2009/01/more_than_codin.html
I was interested to find a stand I've taken for a while now is vindicated by a report from Sandia National Labs. The issue is that of multi-core processors, and the speed increase they give. My position is that current multi-core/multi-processor threaded programming is useful for only a short time, because each time you add more cores you get less and less improvement in speed.
That's because all the cores share the same memory, and it means that you get more chances of processors having to wait while another core finishes using the bit of memory the first core wants to use. This is known as 'contention'. There is also the fact that there is only a limited memory bandwidth for the chips to share.
So while you may get a significant speed boost when you put a second core on the chip, it's not double. Each additional chip you add gives less of a boost.
My solution is to give core its own memory and use message passing as a way of communicating between the cores. In effect you are getting a network on a chip! The paper from Sandia actually backs this up - and what's more the problem occurs a lot faster than I thought it would. My assumption was that 24, or maybe at the most 32 cores would be the break even point where there was no more improvement.
Not so. According to Sandia you get significant increases in speed going from two up to four multi-cores, but virtually nothing when you go from four to eight multi-cores or from eight to sixteen. Even worse, which I really hadn't thought about, the more cores you add after sixteen, the worse the performance!
So, Intel and AMD, over to you.
http://www.sandia.gov/news/resources/releases/2009/multicore.html
Scanner: Other Stories
Is the UK.gov IT gravy train heading for the buffers?
http://www.theregister.co.uk/2009/01/12/it_contract_review/
The Presidential portrait goes digital
http://change.gov/newsroom/entry/new_official_portrait_released/
UK email law
http://news.bbc.co.uk/1/hi/uk/7819230.stm
http://uk.techcrunch.com/2009/01/09/the-uk-governments-plans/
Oracle batch of patches eclipses Microsoft Patch Tuesday
http://www.theregister.co.uk/2009/01/09/patch_splurge/
Nortel hits a debt iceberg; can it survive?
http://newsletter.eetimes.com/cgi-bin4/DM/y/eBOcy0FypUC0FrK0G7WW0EU
USB 3.0 Is ten times faster; get it in 2010
http://tech.blorge.com/Structure:%20/2009/01/07/usb-30-is-ten-times-faster-get-it-in-2010/
Smart lighting: New LED drops the 'droop'
http://www.physorg.com/news151003742.html
*The term 'Ministry of Defence' is a very early example of political correctness. It used to be called the Ministry of War before World War II. As far as I'm aware it's never been called on to repel boarders attacking the British Isles since the name was changed, though it has been involved in a lot of wars...
Acknowledgements
Thanks to readers Barb, Fi, lois, and to Slashdot's daily newsletter for drawing my attention to material used in this issue.
Please send suggestions for stories to alan@ibgames.com and include the words Winding Down in the subject line, unless you want your deathless prose gobbled up by my voracious Spamato spam filter...
Alan Lenton
alan@ibgames.com
18 January 2009
Alan Lenton is an on-line games designer, programmer and sociologist. His web site is at http://www.ibgames.net/alan.
Past issues of Winding Down can be found at http://www.ibgames.net/alan/winding/index.html
|
