Posts by sam 16

Re: RE: VPN usage

I create a cheap US hosted cloud instance on say Digital Ocean, and then use SSH to create a tunnel to it.

On Linux that's just a case of typing ssh (user)@(instanceip) -D10000 and setting firefox to use localhost:10000 as a socks proxy.

On Windows, you'd probably use Putty, instructions here: http://blog.ashurex.com/2012/03/15/creating-ssh-proxy-tunnel-putty/

The advantage of this is that Netflix can't block you, even using IP blocking, because you are not connecting from a known proxy.

Re: Eh?

The patent linked appears to have been filed in 2000, predating, I think, desktop search.

The patent appears to be about how to rank hits coming from a variety of sources, based on how good the hit is, and how likely the user is to want it (do they tend to click email results? did they look at this recently?). The prior art talks about things like google search and the find function on windows, and explains that this is an incrimental improvement.

Honestly, it saddens me to see money that could be spent making the world better, spent on arguing about money. The greatest tragedy waste of the lawers precious lives. They won't get another. Couldn't they spend them doing something a bit more worthwhile?

Anyway, if I think apple has a good shot with this one, certainly legally and possibly morally. It's just intuitively rediculous tha heuristically ranked omnibar results are under patent in 2012. Maybe minor software patents should have a 1 year life?

The cause of overspends usually seems to be Changes To Requirements after the inital, competitively tendered contract is signed.

Try that with your builder - half way through building a home extention, you tell him it needs to be half the size because you can't afford it anymore, and have a toilet. He will look at the 3 foot high wall in your garden, and charge you a lot of money.

You talk to people working on major, overspending contracts, and they will tell you every time that the customer insisted on changing something, and now it was going to cost a bomb.

Re: "Samsung is on attempt number 6 and they still failed"

I got trolled really badly on this thread actually.

After not reading the reg for a while, I took the comments seriusly.

"What a turn around for MS, people actually buying thier phones"

"And for Samsung too, I though they were gonna be the new apple'

"Oh...wait..."

Re: Piracy = Prosperity?

There is a perspective that the development world uses IP and copyright to keep the developing world down.

Want a nuclear power station? Better pay the French.

Want GM Soya? Better pay the Americans.

Want software? You will be paying the Americans again.

Can't afford to pay because your yearly average wage is 2k, and the software is costing a quater as much as the employee? Well, I guess you can't have computers, can't have the productivity multiplication that modern industrial control, comunications, and logistics brings.

But you can have all these things. You just need to not enforce copyright. You can even use your state agencies to steal western industrial secrets. Some fools will hand you blueprints and code as part of a government procurement contract. Especially if they need thier nuclear power station saftey checked by your agency. The new, native chinese nuclear reactor design looks a lot like the French reactor they bought a few years ago.

You could suggest a lot of Chinese growth is simply 100 years of technological development in fast forward, without having to invest money to develop or purchase the tech. But to the Chinese worker, earning several times what his father made, or the Chinese politician, worrying about how to stay in power, needing to deliver growth at all costs... why does that matter?

Re: So what you are saying

That's a perspective, and I can understand it. Personally, I couldn't do a job well with that attitude because I'd spend my day waiting for 5pm, no later.

There is another perspective, and I think it counts for a good few readers.

I enjoy working in tech. I'm a programmer, and on my weekends I like to program too.

Having a job is wonderful, because I get to program, and people pay me. But the pay isn't the foccus - I'd be coding if I were unemployed.

That isn't to say I don't understand the money part of buisness. It's enjoyable actually - it gives you some measure of your success above nebulous things like how happy your users are (I mean, that one fat stakeholder always complains anyway right?). Making a massive cost saving that increases profit on a job, or landing another project is a great feeling.

But I don't think your adversarial approach to the above is good. I don't think it will be effective, in the long term. That supplier you screwed over won't want to work with you again. A good supplier is a valuable thing, and getting let down by a new, unreliable one is blooming expensive. As for increasing your cost to a client, I guess if your client is the government they might hire you next time, otherwise not so much.

I dunno, I guess I'm saying that you don't want to do a good job, although you think you do. You think you are increasing your employers profit, but actually you are gaming your next performance review. But not the one after, when they spot the patern.

Re: Personality Datastructure

You are right that Facebook is successful because it's awesome. Arguably, windows achieved it's monopoly position by being the best product on the market. However, these products become very hard to compete with due to lockin. Even if G+ has better features, very many of my friends use only facebook. Facebook does arguably have a monopoly, enjoying some 95% of social network usage time in the US.

The value of a social network is in the friends you have on there. That makes them uniquely able to see off competition through pure momentum. This is bad for the consumer because it means no one can win against facebook by inovating. It could be broken by forcing them to open thier APIs in a massive way, so that a post, and a profile, on G+ was also a post on Facebook for instance.

I hope social networks will decentralise, so that we can have a choice of supplier, more like email. I don't think this will happen without intervention. I think this is due to momentum, due to monopoly marketshare. I think facebook is a bundled product (it has many components, you don't get to choose them). I think there is room, and utility, in a slap on the wrist here.

Re: This is new?

A fascinating plunge into traffic sensors!

These days, the primary detection method is inductive loops. A long time ago, rubber bars with a hose in, a more permenant version of the sensus point, were popular.

These days, more advanced sensors are getting common, for instance radar sensors that can monitor multiple lanes without having to replace on resurface. Also infra red sensors for more accurate vehical profiling, height etc.

Another recent inovation are little wireless capsules, containing an electronic compas, that detect car type/speed/count based on the vehicals disruption of the earths magnetic field. The highways agency trialed them a couple years ago, and found them much faster than loops to fit, but battery life in field remains to be proven.

Another exotic technology is multi microphone sensors which use traffic noise to give a car, lane position, speed etc. Less accurate than loops but cheaper.

The UK uses automatic traffic control in cities. The individual lights all report back to a central computer which optimises traffic flow in quite a deterministic way.

It is much cheaper to link the lights to a central station because you can just use BT phone lines.

Of course, you could do self organising by linking the lights via a central hub at the same cost, but if you do that, you lose the redundancy anyway, so there doesn't seem much point.

IMHO, real progress could be made by harnessing data from peoples GPS devices to let you know where they were trying to go. Perhaps you could send instructions back to the GPS, routing people evenly accross varius possible routes. If this happens, it will come from the GPS/mapping industry, not the traffic lights people.

Re: Same old, same old

Why does microsoft care about IE?

A long time ago, and far far away, IE6-only web apps prevented migration to other platforms, but now IE is near compliant, and multi-browser libaries are pretty common, it just doesn't have the same lock in factor.

What keeps devlopers on MS platforms now are the awesome dev tools, and the user base. Cooperate IT love it because the network administration tools are built around use cases they actually need, and useful, rather than the roll your own linux approach, or big fat nothing in the case of tablet platforms.

Users stay because it came on the PC, and because it's the platform developers write for.

Where exactly does IE fit into this as an asset? Surely they just need to write 'a browser' to tick of that feature, and then hope someone writes a better one?

MS do not make the best antivirus software in the world, but there are so many 3rd party offerings it doesn't matter. Same with image edit

Re: Money vs. quality

As I understand it, the problem is that he lied on SEC filings - his report to the market about the company. That's kinda illegal.

SEC filings include other information like profit and loss figures. If you can't trust what is written there, stock markets can't work. It's kind of important for capitalism that we are serius about enforcing accuracy.

The later part of your question is more whether we should want it to.

Activist shareholders have a bit of a record as asset strippers. You buy a stake in a company, and maybe borrow the money to buy it. You need a return quick. Fastest way to do that is to seperate out the profitable parts of the company, detatching them from scary things like pension commitements, and sell them off. The parts becoming worth more than the whole. Profit.

Is this a bad thing?

Yahoo has some great assets. Which would have a brighter future as part of a different company. Yahoo Japan would be an excelent purchase for a Japanese TV network or mobile carrier.

As part of a company that is unable to invest, or just really bad at internal investment, that unit has a sorry future. It will support the worse parts of the company and not make the great products it could with another owner.

Asset stripping is pretty horrible when it happens to your company, but can be a good thing in the long run, if you are in a good unit. Bad ones are parasites, kept alive artificially by being in a big company, better that they should die.

Re: out of interest

The intention of an advert is generally not to sell you a product, even if it appears to be advertising one. The intention is to make the brand visible, and assosiate it with the product. So that when you want a netbook, the word Samsung pops into your head, because you have seen samsung next to netbooks out of the corner of your eye many times.

You do not have to read the ad for it to sink in. I know the ad on the right is for microsoft, even though I haven't looked close enough to see what product it is for. It has a man in a suit. He looks responsible in a rugged sort of way. Look at his confident expression. Your brain recognises symbols (the MS logo), human charecteristics, and colour, without concius thought. Thinking buisness? Think microsoft.

Nintendo spend twice as much on advertising as they do on development (software and hardware). Why?

Because price is set by Supply and Demand. R&D creates supply, and Advertising creates demand, it directly effects the price you can charge for the product. We have no way of knowing a true price for most things we buy, which is really better out of 2 similar spec laptops. But we will pay twice as much for Microsoft hosted office web app compared to some nobrands feature matched web app. Because we care about uptime. Because Microsoft means buisness. Because the guy in a suit looks like he means buisness, even if we never really noticed him.

Re: Where's a Mr.G.Fawkes or associate when you need him.

I don't think it would help as such, but you have to admit it would be kinda awesome...

Re: R C-J is probably the right man for the BBC job, but has overstretched himself here

Well, it appears that he wrote some javascript, partly by reading tutorials and code from other websites.

I don't understand exactly what else you folks were expecting.

What, did you sit down to write your first web app and Just Know JQueries API without looking at some examples?

Did you not paste those examples into a browser, then poke them till they did that thing you were looking for?

Did you not take these modified examples, and nit them badly into a web abomination?

And then sit back and go 'Cool, I did a bit of Javascript. And it was easier than this DIV based layout thing in HTML'.

So Rory didn't channel the spirit of Stalman. Is he even dead? No? Oh well, hopefuly soon...

But an ARM is specialized...

ARM chips are RISK processors, specialized towards flow control operations and simple arithmetic. They use pipe-lining the push a lot more operations through per cycle than the CISC chip you get in your Desktop.

But they are rubbish at the kind of high throughput mathematics that is required for video decode, and even wireless networking these days. CISC chips have massive instruction sets, giving access to a combination of DSP hardware and optimised microcode for vector math. It's not as extreme as a vector engine, but its there.

For me, this proposal that packages should contain a range of semi-specialized hardware to carry out different types of generic computing task is a migration back to CISC, a surrender of the RISK concept that has dominated mobile devices.

Backwards compatibility has crippled desktop CISC, and I hope that the new specialist CISCs will be a bit more pragmatic rather than being shaped by the migration from previous hardware. A nice way to achieve this would be for SOC vendors to offer a huge base of C++ libraries, with the proviso that the instructions set was prone to change between devices, and using it directly was asking for trouble...

Good buisness...

Generally people don't buy the best product. They buy the best sold product.

Once one antivirus company is willing to pay for a trial version install, all of the others have to, whether they have the best product or not, because most users will just upgrade the trial version.

If you are procuring for a public sector contract or a big company, the manufacturer assumes you will want to see hard data for why their product is faster, cheaper, has a better detection ration. They go out of their way to make that data available to you, draw your attention to whether their product differentiates, and where it is independently verifiable.

When marketing to a consumer, the companies seem to reduce information content to zero. I see an ad for a credit card at the station - it features a sexy woman, a man in a suite, a catchy slogan, but no graphs. Sad.

Wireless last mile...

Remember that wireless bandwidth is shared by everyone using the same channel.

I calculate that, using all licensed mobile band width, 4G will provide about 4Gb/s capacity, shared by everyone in your cell. The real maximum will be much lower than this, effected by interference and multipath.

4Gb/s sounds like a lot, but there are currently about 5000 people a cell, so you'd get less than 1mb each.

This can be mitigated using much smaller cell sizes with lower transmitter power, but at some point you run into trouble getting the signal into buildings.

Personally, I think that a network of directional transmitters / receivers, like the microwave network already used to ferry data around the country, would be a great way of avoiding all that digging.

Rather than, y'know, using slave labor.

Cutting a kids food money because their mum doesn't want to do hard labor so I can watch porn while telecommuting seems a bit disproportionate. And that is ultimately what this is.

Keeping the riff raff out?

I think it seems a lot because we see porn free. At least I do.

The people that make it don't though - they see a 20 minute video as costing $10k in 'actors', camera crew, equipment, makeup artists... 10 videos like that to launch the site is a hundred thousand.

For pirate... sorry user uploaded streaming sites, it might seem like more, but I'd guess if they are popular they make a bomb.

But...

It's typical terrorist practice to anticipate emergency service response to an attack, and blow that up too.

In the past, several groups have guessed what hospital the casualties would be taken to and bombed the busy emergency department.

As there were several near simultaneous explosions around the city, expecting an attack on rescuers was basically sensible. This might have included blowing up the hospital, the evacuation route, or just setting a secondary bomb to explode half an hour after the first one.

When it comes down to it though, it's about risk. A firefighter needs to enter a burning building to search for casualties (10% chance of finding one who lives say). At what level of totally guessed risk do you not ask him to enter the building? 10% chance of him also dieing? Seems a tough ask...

Blaming the user?

An email arrives in your inbox.

It's from your brother.

Title is, "Plans for Mum's 60th"

There is no attachment.

The email asks you to have a look at a hotel you heard him mention before.

You google for the hotel, and look at it's website.

A javascript executes quietly compromising your computer.

Your brothers yahoo account had been compromised, and the attacker found out about your mothers birthday from his facebook. The hotel site was an unpatched wordpress installation by the managers son, and the javascript was inserted into it through an SQL injection.

I hope that you read emails from people you know that are about stuff you expect from them. Sometimes even emails that have attachments. Otherwise... why do you have email?

Blaming the user is fine if no one wants to specifically attack your business. When you are a government lab targeted by professional intelligence agencies, I'm not sure it works.

What customers?

A lot of companies would love to use thin client PCs, so long as the 'cloud' could be a server room physically on site. The Google vs Microsoft case implies that Microsoft is quite happy to offer this - probably users on thin client terminals, a big server room running the 'cloud' software in the customers basement, and data replication to other customer sites and a big backup center.

Easy to support, doesn't rely on your data line, centralized control of information...