« October 2006 | Main | December 2006 »

November 28, 2006

The Worst Vista Feature

No, it's not the coarse-grained User Access Controls that people are likely to turn off because it annoys too much.

Instead, it's the new Disk Defragmenter. I mean, look'it:

vistadefragmenter.png

"This may take from a few minutes to a few hours"

That's not only crap English, it's insufficient information. I want to see coloured bars, little cubes, progress thingamajigs and spinning stuff. And, I want a report at the end.

Update You could do it manually, with Administrator privs, comme ça:
defrag.png

... which gives you a progress report at the end, but nothing else. The Resource Monitor doesn't provide anything blinky-blinky either for the defragger. :\

November 26, 2006

"Avoid the loony Zune"

Andy Ihnatko's total and utter destruction of Zune is an excellent read (yes, even if you happen to like Microsoft's new media player):

http://www.suntimes.com/technology/ihnatko/147048,CST-FIN-Andy23.article

Microsoft's colossal blunder was to knock the user out of that question and put the music industry in its place.

I'm guessing there won't be very many sent out to competent reviewers any time Zune...

November 23, 2006

A virtual stadium for Auckland

The crocodile tears, court injunctions, media beat-ups and sheer bickering over the Auckland Waterfront Stadium are simply unbearable. Something needs to be done.

Clearly, there will never be an agreement to build the stadium because it's not crass enough and people worry about the money even though it's not actually that much, despite the wild $1.6 billion guesstimate from Burqa Bob the Builder in Tauranga.

You could argue that it's embarrassing that there's no world-class stadium to host the RWC in Auckland (or any other sport for that matter) and it is. Then again, Auckland likes that kind of embarrassment. That's why the public transport here sucks and we don't even have a motorway to the airport for instance.

It does bother me though that nobody's hit upon the obvious solution for the Gordian Knot that is the stadium: use technology, and make it virtual.

That's right, virtual. It could be hosted in Second Life or if Auckland put a little bit of effort into it, a similar virtual world developed locally. The entire country could then join in and design the Virtual All Blacks and best of all, it wouldn't all have to go through Sky either.

There'd be plenty of excitement too, when say the French decide to hack the VABs because that's the only way they'll ever win. Betting on the games would be easy too and there'd always be a seat at the stadium for you.

It's not as crazy as it seems eh?

November 20, 2006

FreeBSD 6.2 RC1 available

beastie.png

I'm a happy FreeBSD camper, using 6.1-RELEASE at the moment as my main Internet server and gateway. It's stable, fast, easy to maintain and Just Works for what I want it to do. The next point release, 6.2, is coming up however, and needs more testers for the first Release Candidate.

If you're thinking about trying out a powerful and secure Open Source operating system, you could do a lot worse than FreeBSD.

An NZ mirror to replace the existing one on Citylink is in the works - will post the URL once it's up. Here's the official RC1 announcement:

We have now reached the Release Candidate stage of the FreeBSD 6.2 release cycle. A few significant problems had been discovered during the initial BETA testing and those issues should now be fixed. RC1 is the first of two planned Release Candidate builds. If no more significant problems are reported 6.2-RELEASE builds will be done after RC2.

This is the first point packages have been included with the test builds
and a few minor nits have been noticed already. In particular these are
known problems which will be addressed before RC2:

- sysinstall tries to install the wrong package for Linux
emulation on i386 if you try to install it when prompted
but the correct Linux emulation package (linux_base-fc4)
is included on the distribution media

- if you install gnome2 off the distribution media it will
fail to find packages for gmime and gmime-sharp

- RC1 distribution media does not include the documentation
tree, RC2 will include a separate docs CD. The disc2
image will just contain packages

We appreciate all the testing and feedback people have been providing,
it has helped improve what will become 6.2-RELEASE. Your continued
testing would be greatly appreciated.

To get RC1 you can download the installation media from the FTP mirror
sites as normal. For those of you who would like to update an existing
system using cvs or cvsup use RELENG_6_2 as the branch tag when updating
your source tree. Any problems can be reported by submitting a PR
and/or sending email to the freebsd-stable@freebsd.org mailing list.

MD5s and SHA256s for the ISOS:

MD5 (6.2-RC1-alpha-bootonly.iso) = 8e39ee88c17884badc3fe189c6ed1795
MD5 (6.2-RC1-alpha-disc1.iso) = 25157cbebdb9820d37dd69131c559c1a
MD5 (6.2-RC1-alpha-disc2.iso) = a482d215c05a24df55e5e263ef77918e

MD5 (6.2-RC1-amd64-bootonly.iso) = 1185b38c1ea6fc93f3d1dffb0d40cba9
MD5 (6.2-RC1-amd64-disc1.iso) = ec29dae6926c1641a6cc7dd1a2047f84
MD5 (6.2-RC1-amd64-disc2.iso) = d9de53123d42a6464ae6d8a930024f80

MD5 (6.2-RC1-i386-bootonly.iso) = 2d0ca001f27e342aaffb265ca65fcde2
MD5 (6.2-RC1-i386-disc1.iso) = 29887fdc63ca47b60febe98a7246c451
MD5 (6.2-RC1-i386-disc2.iso) = bfc991723d29bb320db302b47871b1f9

MD5 (6.2-RC1-ia64-bootonly.iso) = cf6f1519b4dd4264fb2b873bc280a8f4
MD5 (6.2-RC1-ia64-disc1.iso) = 983e795f65882bfd93250ec074b33ca6
MD5 (6.2-RC1-ia64-disc2.iso) = 4699d6f27075144cbe2347271d47c89c
MD5 (6.2-RC1-ia64-livefs.iso) = cdb6c5d853acaad1d8385d4c8302d061

MD5 (6.2-RC1-pc98-bootonly.iso) = 43df2d851438b5212faa81e456190ae5
MD5 (6.2-RC1-pc98-disc1.iso) = ba33bc96d1c416f16e123d43c88110da

MD5 (6.2-RC1-sparc64-bootonly.iso) = dd29f3da4efb23d84fbb051db556b62e
MD5 (6.2-RC1-sparc64-disc1.iso) = f4f4025cbb68cf332ccafb44a4973ce6
MD5 (6.2-RC1-sparc64-disc2.iso) = 8ce06a2a633141cc5a52f8d1f6059e22

SHA256 (6.2-RC1-amd64-bootonly.iso) = c3b0e1192f0756bf36438e4885d312cec03da816e3cf3682334ac627eec59713
SHA256 (6.2-RC1-amd64-disc1.iso) = 1b83029274ca3aac1a9304abf2d9a97d1a6020b01066385ed7378826ce00517c
SHA256 (6.2-RC1-amd64-disc2.iso) = 0ec1a6562b92e917a77e0a536cc31c6a510ae77fcd58369342a3cfa69d61aa45

SHA256 (6.2-RC1-i386-bootonly.iso) = fdabfbf8b60c21a37d0311f89cd72070c080a7985cc7fd588a1365cbca6d24bf
SHA256 (6.2-RC1-i386-disc1.iso) = 1609f2ddf3afd353f2451f36e9554b332a67ff1539e90d0d2023aee21b0ab513
SHA256 (6.2-RC1-i386-disc2.iso) = f49fd1ce0769199eeb840046817249d96c6c95631b5546586d46d6cb5d760640

SHA256 (6.2-RC1-ia64-bootonly.iso) = 51f410cfab7846fca7576e265dd1683f1494f71560714c17fe7f9f0695389041
SHA256 (6.2-RC1-ia64-disc1.iso) = fc39dbe8e99766a7b071fa35f4efea6a65728d05379ca29cbefec8431901b868
SHA256 (6.2-RC1-ia64-disc2.iso) = 3119d0d625a6abb218653c339976d9a6b84272d71f94573f06a26fa070ed97cc
SHA256 (6.2-RC1-ia64-livefs.iso) = e6c8411c406c497c2bc5ad9577f2db159a3c8a39ef3123674faede9765c08107

SHA256 (6.2-RC1-pc98-bootonly.iso) = 66d30a5a3ca1b296cf9370a9116a1cce282d5f25d1843be36c21d731b4225bf6
SHA256 (6.2-RC1-pc98-disc1.iso) = 1a1ff42156d8e201feb54c6a1e36c3b07a682b22fc7ba8a13fee0ebef2316c26

SHA256 (6.2-RC1-sparc64-bootonly.iso) = ef493ca3b5f053cfcc10f9f5f0d7e85171681fe2f1aa67d314b63a8bc8d25983
SHA256 (6.2-RC1-sparc64-disc1.iso) = a2a3ac1191e7ce46efb12d1c4e51a0eed5a72d130f7d48affa0f8fcfa1324c41
SHA256 (6.2-RC1-sparc64-disc2.iso) = 79103893f4059fca06bb20a09df157ff28c814fa33f5db715c6b87a81eb0711b


--
Ken Smith
- From there to here, from here to | kensmith@cse.buffalo.edu
there, funny things are everywhere. |
- Theodore Geisel |

November 16, 2006

Zuney stuff

It is quite interesting to see how Microsoft's become so large that it's starting to fragment into mutually incompatible pieces.

You can spot some of this in the Live/MSN dichotomy but nowhere is it more apparent than Microsoft's new media player, the Zune.

Microsoft has already made forays into the mobile music market, with the Playforsure DRM stuff, the MSN Music Store and Urge offerings. Zune isn't compatible with any of them.

Where it gets comical though is that Zune isn't even compatible with Windows Vista currently. The plan is that it will be by end of January when Vista launches, but it's not a good look for Microsoft.

Speaking of Vista, I've taken the plunge and started using it on a daily basis now. Still on RC2 but should upgrade to RTM soon - no, not Zune. Little bit disappointed in the fit and finish of RC2 actually. The OS itself is rock-solid and runs fast, but I get a lot of application crashes. Most seem to happen when Vista goes to sleep and it affects parts of the OS like IE7.

IE7 in Vista is also curiously different from IE7 in XP. Some sites don't render right in Vista, but look fine in XP - Google Gmail is one. In fact, I'm having a few problems using Movable Type (the blogging engine that PC World uses) in Vista IE7, which doesn't like "scriptable windows". Oh well, Firefox 2.0 runs fine on Vista, ditto Opera 9, so I'm not too fussed.

November 14, 2006

Maxed out broadband not the government's fault

Editor Chris Keall's been talking to Stephen Crombie at Telecom and it's interesting to note that despite all the promises of playing fair, openness and going along with the new reality, our beloved incumbent still tries to confuse and obfuscate.

Crombie says:

The Commerce Commission-enforced Max experience was never going to be good for everybody, Crombie said (contrary to expectations raised by the "Xtraordinaries"), because essentially the Max plans mean switching off the network management controls that evened out everybody's bandwidth. Now, those who live further from exchanges will suffer more.

Let's debunk that "Commerce Commission-enforced Max experience" item first: Telecom didn't have to provide unconstrained commercial service to its retail customers. In fact, it was only required to provide regulated UBS to Callplus and Ihug - and that particular service is only unconstrained on the downlink. The upstream is still only 128kbit/s, as per the Telco Act.

The Commerce Commission never told Telecom to take off the rate-limiting on its commercial retail and wholesale DSL.

It would be interesting to know what "network management controls" Crombie's referring to. Telecom never bothered to put in an interference management plan when it rolled out its DSL network, so it can't be that. If it's just the rate-limiting, then we're getting closer to the root cause of the performance issues: namely that Telecom's network doesn't have the capacity to handle a still small, but growing number of broadband customers.

The issue about customers living further away from exchanges having to suffer because Telecom is now offering DSL without rate-limiting, something other telcos around the world manage to do without problems, was dealt with by Paul Brooks of Layer 10 (link goes to PDF file with report).

Brooks' earlier report to the Commission found that:

... ”bit-rate limited„ ADSL1 services would provide little benefit over unconstrained services in terms of their impairment of surrounding ADSL services, particularly in the scenarios under debate at the time. In particular, there was no additional risk to marginal services on very long lines from unconstrained services - additional unconstrained services would provide no more degradation than constrained services, and any detrimental impact would be due to the increased numbers of services of any form, not whether they were constrained or otherwise.

Telecom released a report by its network manager Alcatel that predicted dire consequences if "unleashing" took place - and you'd have to wonder why then if Telecom isn't mandated to provided unconstrained broadband, and if it is so detrimental to DSL reach and service quality, it would even think about providing "Max" plans? Could it be that the report's conclusions were bogus?

This is what Brooks says about it:

In summary, while the observations certainly seem to indicate that increasing numbers and speeds of ADSL services (of any variety) do cause increasing levels of interference, there seems to be no evidence that unconstrained services are likely to cause any more impact than similar quantities of constrained services (constrained either by bit-rate limiting or reduced transmission power). That increasing numbers of ADSL services can cause degradation of adjacent services is well understood and accepted. Telecom and Alcatel must, however, show that unconstrained services will cause significantly more degradation than the same number of constrained services - and this is not a conclusion that can be drawn from the analysis presented.

Well, yes, it really does seem that Telecom's claims are rubbish. How would Telecom deploy ADSL2+, with its far higher bitrates (24Mbit/s down, 3.5Mbit/s up with the latest Annex) over such a network?

If Telecom was really concerned about customers living a long way from the exchanges, it would deploy ADSL2 Reach Extended for them. This would be quite cheap to do as well.

The only services at risk from increased interference due to greater customer numbers over the copper network are the high-value business G.SHDSL ones - and that's because Telecom doesn't have an interference management plan.

All in all, it's wrong to blame the Commission for Telecom's blatant underinvestment in network capacity. Without regulatory action, you'd still be on 128kbit/s Jetstart, with "innovative new surf your inbox products" like the 64kbit/s plan Xtra mooted a few years ago. That one didn't have any data cap, only 20¢ per MB charging. Ideal for people just wanting read their email and do a bit of web browsing apparently.

November 3, 2006

Microsoft embraces Novell SuSE Linux

And... Microsoft may well extend it too. We'll see, but either way, today's announcement is pretty big.

Microsoft says the key focus is on providing intellectual property reassurance, but the deal is also a gonad kick for Oracle and Red Hat.

See my Computerworld story for more details.

November 2, 2006

PC World for $5 on Trade Me

Cheap and good! Buy now!

pcwtrademe.jpg