Does it support pipes? Maybe we can make C|N>K for real.
90 publicly visible posts • joined 16 Apr 2007
I used to do validation of this sort of data all the time, there is a boatload of tools and the methods are well published (though it seems not well known). I've seen a great slide illustrating spatial data rejected by a validation frameworks, pity I can't find a version on the web but you can tell *what* sorts of errors are the most popular (this particular one is in the top 5). For instance, getting your coordinates out by a decimal place gets you a miniature dataset flowing in the sea off the coast of west Africa near the Greenwich meridian (hint: if it's continental data, it's probably wrong). Once you have a feel for how it can go wrong, you can look out for it. (Cities should be on a land mass unless the name is "Atlantis")
Same for attribute data. You have pH values? Domain better be between 0 and 14. Temperature? Minimum should be less than maximum. You get the idea. If you use data from external sources, you have to be paranoid about QA.
Your privacy on social media can be measured as the aggregate stupidity of your friends. The trend seems to be that the purveyors pester people to upload their contact lists, complete with names, mobile phone numbers, email and real world addresses. If you are flogging a mobile phone as part of your empire, you probably don't even need to ask. It would not take too many slupred mobile phones to get nearly all the details and some statistical rigor on the reliability.
They devices won't just snitch on you. They will coddle you in sometimes illogical and frustrating cotton wool. Remember the Java JINI kitchen joke in 1999?
"You go to put your jini milk in the jini fridge and for no apparent reason (some software conflict probably) the fridge refuses to accept it. No matter what you do the fridge just keeps telling you there is no milk. So you just decide bugger it, it's still keeping the milk cold, but then when you go to make your breakfast porridge the microwave won't turn on because the fridge has told it there's no fresh milk. Neither will the element on the stove work because the house has determined you are trying to cook up spoiled milk. "
@AC 00:05 What have the Romans ever given us?
While the Romans probably didn't invent the aquaduct, they certainly made huge advances in making its benefits easily available to lots of people.
Sort of how MS took a bunch of Unix technologies as made them available in a ready-to-roll form as Active Directory. Before that it was their acquisitioon of MS-DOS.
Seriously. It is well known that important transactions need to go through more than one person so corruption requires the complicity of more than one person. One disgruntled employee? Very possible. Two? Much harder.
How hard would it have been to have the form/transactions tee-ed off to a second senior plod rather than a DBA?
So many simple precautions could have made this a nuisance rather than a data loss.
My coat is the one with the Glock.
So currently guns are fundamentally mechnical systems and have no need for biometrics or electronics to function. I can't see how you can rig up something that cannot be removed by someone with determination and time. The main instance where this might help seems to be snatch and grab situations like on the street (where I'd hope that most private owners aren't gadding about packing heat anyway).
And we've not had a resounding success with DRM of fully electronic systems either.
But like cars, we may see more electronically integrated firearms, maybe with electronic firing. As this becomes common, we might insist on such firearms being used in circumstances where snatch and grab might be an issue.
Like a number of people have said, it's just an appliance. The vast majority of people who will buy this have no interest in buying it, they will probably see some reliability benefits from a more solid homogenous design (ala potted electronics). Those people who want to be able to upgrade will buy something else. Sort of a budget halfway house between an iPad and their other desktop offerings.
Asides from the space junk, there are some wonderfully intersting problems in the collaborative data space. Enough for a good PhD even.
For instance: 1) Two people map the trees in their local park. One person did it several years ago and paid great attention to accuracy and detail. Some years later a second person maps the same park, while their map takes into account all the recent changes, the collected data was rushed and is not as accurate. 2) I use my GPS to map dangerous rocks in a local bay, what can we say about the absence of dangerous rock markers elsewhere in the bay?
A lot of this is covered in established theory (like sins of ommission, sins of commission), but collaborative data adds some nice complications and scale problems. My own view is that with enough data and understanding of biases in data sources, we can average out the real answer.
Most people who really care about their email and privacy are unlikely to be using their ISP's email facility anyway. Email is a commodity, I can see why Telstra mightn't see much advantage in maintaining their own inhouse bespoke solution (that penny pinching customers end up paying for the maintenance of anyway).
In some countries, for the price of a PVR you can subscribe to a web based service that records all the free to air shows for you in a PVR like manner. People I know have adopted it because the annual subscription is well worth the hassle of not having to remember to record/download/store things. Presumably content providers get a slice of that revenue. As far as I can tell, everyone is happy.
About the only thing it doesn't do is stop you from ripping off the advertiser by going to the loo in the ad breaks.
This article in El Reg last November does a great job of diving into the technical nitty gritty of Neal Young's grievances. Worst thingis that things like widespread ignoring de-emphasis flags gets people used a particular forms of distortion to the point that audio as Neal intended might "sound funny".
Brad Fitz's Social Graph was (is!) actually an awesome idea, but it's perhaps no surprise that the Social Graph API is a casuality. The Facebook horse bolted a long time ago and there's too many monopolies at stack to get all caring and sharing. This is a real pity and still worth a look as it does point towards what social networking *could* have looked like.
Forget the side issue of masses of questionable quality home brew textbooks that won't ever show on the marketplace radar anyway.
The publishers of textbooks should be relishing this (except for potentially losing control of the means of distribution). Can you ever see anyone allowing any sort of second hand textbook market in this scenario? Bypasss the doctrine of first sale in one easy step, go straight to "Go" and collect $200 (for each copy, preferably via an annual licensing charge).
May as well veto other automotive distractions like car radios while you are at it.
Truth be told, the GPS on my phone has entirely eliminated the need for me to lean over and find my place in a dead tree edition street atlas while driving.
Nearly all our blue label number crunching software has been charged on power units, CPUs, cores and sockets. All of them have had to change to some degree to accommodate VM farms and clouds. The responsiveness seems (to me) have been driven by just how close their nearest competition is.
What's also interesting is the number of stats grads coming out of university pre-loaded with R. A quick browse around also shows a lot of work to link R with things like Hadoop.
This is an entire discussion that would have not even been possible if this compute capacity had still been scattered across servers and desktops the world over. I wonder if this is just so much like Richard Pryor in Superman III where all these fractions of cents add up to a really big number. Now we need a context in which to understand Google's cinsumption of electricty.
"The WASP's compact computer may of particular interest to News Corporation since it is capable of sniffing Wi-Fi networks and intercepting mobile phone calls. "
Well, yes. You could probably also attach cream pies, weapons of mass destruction and an electric shaver to it if you really wanted to. So what?
There's a smart phone mania gripping all OS developers who assume that you now have a tiny 3" x 4" screen andd cannot see more than a playing card sized chunk of document at a time. I gave single task foicussed Gnome 3 a red-hot go and just ended up hating it even more. I notice Lion is enthusing about the same mis-features that annoyed me in Gnome 3. Windows 7 takes a POV if one user action is good, three actions are better (here's snarling at you ribbon bar).
"GNOME emerges from last century"... and assumes I have a low res 3" x 4" dsiplay device and only ever want to see one thing at a time.
I have been giving Gnome 3 a red-hot go, but the whole assumption that I'm driving my user experience form a tiny touch screen is really starting to grate. Why shouldn't I be able to have my power settings and network preferences open at the same time?
So if we discount the Nokia N900 and progeny as a long term option, what Atom based handsets can I consider? A quick Google just shows a bunch of excited announcments from 2009 or thereabouts and XMM chipsets being released by Intel. Is anyone making phones?
Yes, I'd love to consider a smart phone that didn't require me to chose between Page and Jobs.
" Yes. It takes boatloads of disk space and buckets of RAM. Objective reviews have shown it to really not be much better than Vista in this regards, it's just that (high end) system specs jumped enough in the intervening years to make Windows users not complain about the bloat as much."
That sounds like every Windoiws release I have ever known (and that's going back a ways). The software has always been somewhat ahead of the hardware curve.
If there is any sense, this should become the norm for television viewing.
- Why should I arrange my life around a TV schedule predicated on single sequential channels?
- Unlike home brew torrents arriving from overseas, advertising is more likely to be retained and seen.
- Better targeting and stats for the ads that are seen as compared to broadcasting to a guesstimated audiences of unknown engagement.
- How many people are already making multiple copies of the same shows with their own PVRs anyway?
Done right, this can rely on laziness and make it easier to pay a few bucks for a subscription and be subject to some adverts than mucking around with torrents. Everyone wins except perhaps the broadcasters. It's a pity we've just sunk a bomb into digital broadcasting.
So I thought the new model was to outsource R&D by risk using VCs to fund startups which then get acquired if they show any promise.
I supopse the art is knowing how long to leave companies like Fusion-io and Virident to reduce risk and grow market before they get gazumpped from under your nose or steal a march on you and trun in to competitors. Then there's that whole adbdication of control (or is that just the ultimate expression of faith in the market?).
The current hierarchical approach has single points of failure that multiply consequences the further up the food chains you go.
I should have a number of certs for my PKI that are vouched for by the people that I know. This would mean you would see a number of mutual "friends" who all (hopefully) agree that I am who I am and hopefully highlights any anomalous outliers. You could also look at the certs accepted by your more technically inclined friends and compare them to the ones you are being asked to accept.
Of coursde this would involve throwing away a perfectly good monopoly and fewer sheeple.
I ran up a new FC15 dev machine last night and it was a case of "same old, same old" until this whacky new UI popped up.
While I'm giving it a red hot go, I'm with Rod on the need for a UI that meets *my* needs. The iOS style UI will suit many casual users, but I need to be able to launch/access temrinals and I don't wwant vanity effects on GUI events. I have not yet looked at whether I can alter settings to better suit my needs.
We would not see any progression in GUI without real attempts at change, but I wonder if any of them are all that much better than the OpenWindows environment that I still quite like. I'll give Gnome 3 a week or two and then make a decision (because unlike other OS, I can make a decision).
Yes, you could design a nuke plant to completely withstand a massive tsunami, but it seems to me the designers figured it was pretty much going to be trashed by such an event anyway. This blog (now picked up by MIT) sets out how things are supposed to work and the observed issues and responses seem to match what the designers said would happen in this sort of exceptional circumstance.
I suppose they could have built the plant up in the mountains, but maybe if they'd done that, there would have been a massive avalanche that swallowed the plant whole instead. Monday's Experts and all that.
I was impressed by mum's commodity telco. Her new ADSL modem came with a proper hard password already set and a refernce card with said password to be stored in a safe place.
It would not be hard for even the cheapest modem to make a password reset mandatory as soon as you log in to connect to your ISP.
The specific refeernce to GPLv3 is interesting. Two key motivations behind v3 were software patents and DRM; two topics that ought to be dear to MS's heart. Methinks that the veto is based on specific fears on one or both of these fronts. Or it may not be in response to a perceived threat, but a measure to impede adoption of GPL v3.
This can work quite well if Nokia is eventually run rather like MS's X-Box division. Bootstrap some mobile phone expertise; Ride the user base long enough to get a toe into the market; Have a preferred primary platform that avoids dealing with a menagrie of hardware; Run the deal lean enough so that a buy out becomes an unbeatable value proposition for Nokia shareholders. That or watch and learn before MS releases their own hardware and cast the withered husk aside. I reckon it's a great reprieve for MS, such second chances do not come along very often.
Other WP7 particpants should be hesitating right about now; they don't fit into the picture particularly well at all.
Further to my prior post, it appears that there is a few people who see the acquisition stars lining up:
"Nokia stock is down at the moment. I wouldn’t be surprised if this also fits well into MS’s plan. They will buy a lot of this stock while it’s cheap, giving them more leverage to influence Nokia down the track."
@ttuk S-Plus and R only share a ayntax. Insightful commercialised the Bell Labs developed S and S+ and added new features. All the beards that wrote S+ went off and developed R. Fact of the matter is that S+ came first and the core product is not likely to draw on any R.
Sort of like refusing to visit the USA because you visited the UK and got accosted by Chavs.
This will be interesting. This seems rather like the $1k per seat market that S-Plus lost to R in the first place. The question is if they can offer enough features to make it a worthwhile alternative to a free version. Or is this a case of taking the whole R phenomenon to the SAS community? I'd wager that far more stats graduates arrive with R than SAS.
This is certainly moving towards how computing resources as a utility is supposed to work.
Now I'm thinking about how even well understood utilities can fail, and the general pattern of less frequent outages with much more widespread impact.
Will we inevitably one day see the Google equivalent of an un-pronouncable Icelandic volcano shutting down more than just a couple of planes? What's the plan then?
There are other FOSS social projects about, some of them with more mature code than Diaspora, some of them with lots of paper prototypes. I think Diaspora managed to get some high profile mentions and got a lot more notice. The attention about Diaspora has also renewed interest in projects like AppleSeed (wow, even a "noun-noun" name).
There is no natural, inevitable social network monopoly. There's a majority social network, but we've always had niches for other social networks to exist. Rememeber that Facebook was originally just for one social group at one university. I don't know if FOSS projects will ever challenge FB, but I can see a number of circumstances in which vibrant alternate social networks will be viable. Also, these small startups won't be as restarined by the consequences of disruptive change.
There was a good analogy drawn elsewhere about how everyone now expects that they can take their telephone number and account details to any telco they want.
This is really serious stuff. FB has a success factor that AOL, MS and Google never had; an outright majority. The critical mass of population held by FB means that unlike anyone else, they can potentially play the exclusion game and get away with it.
Run it just like GMail for now; free, easy and convenient. Progressively make transfer to and from the Internet more flakey and always blame the other party. Maybe put alarming security warnings on email arriving from outside the walled garden. Make external operability a real pain so there's peer pressure for non-believers to get with it and create their own FB email account. I don't think they'd entirely exclude external email, but keep enough support for external email to ward off monopoly accuasations. You want to reach 90% of the world's online population that matter? You'll need Mark's permission.
What's scary is that 99.9% of FB users won't care at all as long as they can get their amusing pictures of cats.
I recall about 10 years ago some guy created a web site called something like "BrandnameSucks.com". After a lengthy legal battle, he was forced to relinquish the domain registration to Brandname. Pretty much the next day, a new domain appeared,"BrandnameStillSucks.com". So how about "istillhateryanair.co.uk" anyone?