I have an iRex iLiad, which is one of the first generation eInk devices from a now-defunct company. Scientific papers are fine, and I've spent many summer days sitting in the park with the sun shining directly on the screen without any display problems. I'm more concerned about the fact that the sun heats up the device beyond what I think LiIon batteries are supposed to tolerate...
Amazon don't have the e-ink market to themselves in the UK any more
Indeed. WHSmiths, one of the largest high-street book sellers, is heavily pushing the Kobo reader, and the cheapest model is about 25% cheaper than the cheapest Kindle.
Google Maps is obviously more than just a map, and the fact that commercial users are so pissed off about the fact that it costs money now proves that there is substantial value in integrating Google Maps, value that they were getting gratis, otherwise they would just say "fuck it" and move on to something else without all the bitching.
The reason businesses are complaining is sunk cost. They spend money developing things using the Google Maps APIs, believing that they were free, and now they're not. Developing with OpenLayers is about as easy and confers the same advantages without needing a licensing cost, although if you're serving a lot of clients then you're expected to serve the tiles yourself, but the software is all free, it's just hardware and bandwidth costs. If Google Maps had been this expensive from the start, then it would not have been a problem - companies would have just not developed things based on it in the first place.
I use OSMAnd. With the free version I need to grab the map files manually, although the paid one will download them from in-app. I currently have maps for northern France, Belgium, and the UK on my SD card, taking up a bit over 1GB.
No, he's demonstrating one of the reasons why Google's revenue per handset is low. He isn't demonstrating why it's lower than Apple's, because he isn't demonstrating why iPhone users spend so much more. My guess is that it just comes down to market segmentation. At the price that someone like the original poster or myself is willing to pay for a phone, there are no Apple products but there are several Android phones. This pulls down the average for Google. It would be more interesting to compare how much Apple and Google each make on average from a $600 handset sale.
What is a PC? A modern high-end mobile phone can drive an HD TV via HDMI at 1080p, can use a bluetooth keyboard and mouse, can have 64GB of flash storage, 1-2GB of RAM, and a dual or quad core processor running at 1-2GHz. Is it a PC, when it's docked at a desk, or in a living room? Does it stop being a PC when you pick it up and unplug the display? What about when you slip it into something like the ASUS Transformer and it gains a bigger (but still portable) screen and an attached kayboard and trackpad?
Samsung just posted $5 billion profit for their last quarter.
Which makes as much sense as you posting Microsoft's total profits when someone claims that Sony is making more profit on consoles, rather than posting the profits of Microsoft's XBox division. Samsung is one of the world's largest manufacturers of flash memory, they make a huge number of ARM SoCs, and a whole host of other components before you even look at their smartphone division, which is a relatively small part of the total company. A lot of smartphones, including the iPhone, contain a large number of components manufactured by Samsung, so they're going to be doing well no matter who is selling the most.
Superior for a lot of things. I'm moving to Cambridge soon, and the university accommodation office uses both Bing and Google maps for their web site (no idea why - it seems quite random which one you get). Neither of them even labels all of the colleges, let along the university buildings. In contrast, OSM labels all of the colleges, most of the university buildings, and even a lot of shops, pubs, and restaurants are there by name.
When I visited a friend in Paris, Google Maps had the street he lived on labelled, but OSM had the building numbers marked as well.
That said, there are a few places where it is less good. For example, it doesn't have integrated route finding, but there are third-party route finders using the same data. If you want to create a map with one marker on it and send it as a link to someone, you can do it via the OSM web interface, but the UI is pretty horrible. If you want multiple tags, then you need to host your own OpenLayers thing and write some JavaScript. The search feature in OSM is pretty poor as well. It doesn't factor distance into account (although the one on the OSM client on my phone does), so if I search for a street name while looking at a city in the UK, I often have to scroll past a dozen streets in random US cities with the same name before I find the right one.
Actually, one of the main reasons that I use an OSM app on my phone instead of the Google Maps one (aside from the fact I don't need a corporate stalker) is that it isn't serving tiles to me. I just grab the data once and store it on my phone. That means I can use the maps with my phone's GPS when I'm out of signal range (or somewhere with only GPRS signals, where using Google Maps is a bit painful) or when I'm in a different country and the data roaming charges would make it stupidly expensive.
The OSM data is licensed in a way that allows redistribution and the project actively encourages people to do this. Clients are allowed to aggressively cache or mirror the data, something which Google or Bing maps do not allow.
Did anyone ever really make a concerted effort to sell tablets to the masses before Apple?
Yes, but the problem was that they were all hardware manufacturers. Apple was the only company that could make the hardware and also rewrite the software stack to be entirely based on touch for input. Earlier tablets ran operating systems designed for either mouse or pen input and most tried to run legacy software. Few people liked them, because the UI sucked. If Apple had shipped the iPad running OS X and existing Mac apps, it would have been a failure.
Alan Kay is always a good source of quotes (including, paraphrase 'I said that 30 years ago! Why does no one ever listen to me?'), but one in particular is relevant here:
If you're not failing 90% of the time, then you're probably not working on sufficiently challenging problems
I think I'd find failing 90% of the time completely demoralising, but it's certainly true that if you never fail then you're probably not exploring really interesting possibilities.
The advantage of the cloud is that it lets you make assumptions that won't be tested until you are unable to restore your backups? Actually, that sounds about right...
I use sshguard, but for a while I was seeing several hundred failed login attempts a day, even though they were blocked after 3 fails (for increasing periods). It may be that after a year my pf table of blocked addresses now includes the entire botnet, or it may be that they just gave up (password logins are disabled on that machine anyway). Or it might just be that the botnet found something more profitable to do than attack a server with half a dozen users...
You're misreading what I said. If you disable the account after n retries, irrespective of where they come from, then that allows a very simple and low bandwidth DoS - just try to log in n times from your botnet and then the real user can't log in at all.
For those interested in a technical / long explanation, see below along with citation links.
Wow. You have way too much time in your hands to come up with such a detailed list. Thanks! I hadn't realised that baud rates kept increasing after V.23bis - I remember reading at the time claims that 2400 baud was the maximum possible with a phone line and that further advances would have to come from other things. The Trellis modulation paper looks like a good way of killing a chunk of boring time on a train next week...
No, it's entirely relevant. The entire point of this story is about people leaving broadcast TV for other ways of getting the same content. It's also the legal definition in the country where I live.
By your definition, every analog TV in the US stopped being TVs when we switched to digital
Yes, if they didn't have a digital decoder box attached, that's correct. They're just analogue monitors then. If they have a set top box that can decode ATSC attached, however, then they are still TVs, because they can receive broadcast television.
The problem is that botnets have a lot of IP addresses. They can do one try from one machine then another from the next. If you disable the account entirely after a certain number of failed logins, you've just created a simple DoS attack. If you disable it just from that IP, it doesn't matter because it will just try from another. There are some realtime block lists that you can use to reject things, but these add another attack route that can let someone who can spoof DNS prevent you from logging in to your own machine...
The mockery was not directed at Palin for being able to see Russia or not being able to see Russia, it was directed at Palin for claiming that being able to see the tip of Russia, about as far away from Moscow as New York is from Paris, had anything to do with her competence with regard to foreign policy.
I've been using DuckDuckGo as my search engine for over a year now. It uses (among other things) Bing via Yahoo's BYOSS API, but it doesn't pass any information about me to them and it doesn't use tracking cookies and works via SSL by default. The search results are usually good enough, and the few times they haven't been I've tried Google and got equally bad results there. The only Google service that I do regularly use is YouTube (which ClickToPlugin makes vaguely useable), and that's hardly something I couldn't live without.
IT's like listening to old people be angry because their VCR flashes 12.
Why not be angry about that? Adding a small rechargeable battery that keeps the clock powered when the mains goes out would add, maybe, 50 cents to the cost of production. Not doing it is just indicative of poor design and lack of attention to detail.
A large part of good UI design is having sane defaults. Saying 'no, it doesn't suck because you can log in and then click on the don't suck checkbox and then it's fine' does not actually make it a good UI.
Because, historically, closed communication systems, once they reach a certain critical mass, have only been displaced by more open systems. A new player can't compete easily an entrenched player because of network effects, but lots of new players can gain control of niches that the major player doesn't satisfy and if they all interoperate then between them they can become as large as the major player - then network effects work the other way.
Google realised this with Google Talk, which is a federated XMPP deployment. On its launch day, Google Talk users could talk with millions of existing XMPP users. The XMPP installed base was probably smaller than AIM or MSNM, but it was already fairly large.
If Google pushed an open standard (hopefully not Diaspora, but something actually designed via a process involving actual thought) then they'd get interoperability with the group of people who don't trust companies like Facebook and Google - probably by now the largest set of people who aren't already on Facebook - and with anyone else who wanted a small slice of this market. Instead, they pushed something that has more or less the same set of disadvantages as Facebook, but without the one real advantage that Facebook has: lots of existing users.
Slashdot Mirror
I have an iRex iLiad, which is one of the first generation eInk devices from a now-defunct company. Scientific papers are fine, and I've spent many summer days sitting in the park with the sun shining directly on the screen without any display problems. I'm more concerned about the fact that the sun heats up the device beyond what I think LiIon batteries are supposed to tolerate...
Amazon don't have the e-ink market to themselves in the UK any more
Indeed. WHSmiths, one of the largest high-street book sellers, is heavily pushing the Kobo reader, and the cheapest model is about 25% cheaper than the cheapest Kindle.
All the years I spent reading paper books, I never once thought "if only this thing had a built-in light."
You can buy them for paper books too. I have one which clips onto the top of a book and lights the current page.
Google Maps is obviously more than just a map, and the fact that commercial users are so pissed off about the fact that it costs money now proves that there is substantial value in integrating Google Maps, value that they were getting gratis, otherwise they would just say "fuck it" and move on to something else without all the bitching.
The reason businesses are complaining is sunk cost. They spend money developing things using the Google Maps APIs, believing that they were free, and now they're not. Developing with OpenLayers is about as easy and confers the same advantages without needing a licensing cost, although if you're serving a lot of clients then you're expected to serve the tiles yourself, but the software is all free, it's just hardware and bandwidth costs. If Google Maps had been this expensive from the start, then it would not have been a problem - companies would have just not developed things based on it in the first place.
I use OSMAnd. With the free version I need to grab the map files manually, although the paid one will download them from in-app. I currently have maps for northern France, Belgium, and the UK on my SD card, taking up a bit over 1GB.
No, he's demonstrating one of the reasons why Google's revenue per handset is low. He isn't demonstrating why it's lower than Apple's, because he isn't demonstrating why iPhone users spend so much more. My guess is that it just comes down to market segmentation. At the price that someone like the original poster or myself is willing to pay for a phone, there are no Apple products but there are several Android phones. This pulls down the average for Google. It would be more interesting to compare how much Apple and Google each make on average from a $600 handset sale.
What is a PC? A modern high-end mobile phone can drive an HD TV via HDMI at 1080p, can use a bluetooth keyboard and mouse, can have 64GB of flash storage, 1-2GB of RAM, and a dual or quad core processor running at 1-2GHz. Is it a PC, when it's docked at a desk, or in a living room? Does it stop being a PC when you pick it up and unplug the display? What about when you slip it into something like the ASUS Transformer and it gains a bigger (but still portable) screen and an attached kayboard and trackpad?
Samsung just posted $5 billion profit for their last quarter.
Which makes as much sense as you posting Microsoft's total profits when someone claims that Sony is making more profit on consoles, rather than posting the profits of Microsoft's XBox division. Samsung is one of the world's largest manufacturers of flash memory, they make a huge number of ARM SoCs, and a whole host of other components before you even look at their smartphone division, which is a relatively small part of the total company. A lot of smartphones, including the iPhone, contain a large number of components manufactured by Samsung, so they're going to be doing well no matter who is selling the most.
Superior for a lot of things. I'm moving to Cambridge soon, and the university accommodation office uses both Bing and Google maps for their web site (no idea why - it seems quite random which one you get). Neither of them even labels all of the colleges, let along the university buildings. In contrast, OSM labels all of the colleges, most of the university buildings, and even a lot of shops, pubs, and restaurants are there by name.
When I visited a friend in Paris, Google Maps had the street he lived on labelled, but OSM had the building numbers marked as well.
That said, there are a few places where it is less good. For example, it doesn't have integrated route finding, but there are third-party route finders using the same data. If you want to create a map with one marker on it and send it as a link to someone, you can do it via the OSM web interface, but the UI is pretty horrible. If you want multiple tags, then you need to host your own OpenLayers thing and write some JavaScript. The search feature in OSM is pretty poor as well. It doesn't factor distance into account (although the one on the OSM client on my phone does), so if I search for a street name while looking at a city in the UK, I often have to scroll past a dozen streets in random US cities with the same name before I find the right one.
Actually, one of the main reasons that I use an OSM app on my phone instead of the Google Maps one (aside from the fact I don't need a corporate stalker) is that it isn't serving tiles to me. I just grab the data once and store it on my phone. That means I can use the maps with my phone's GPS when I'm out of signal range (or somewhere with only GPRS signals, where using Google Maps is a bit painful) or when I'm in a different country and the data roaming charges would make it stupidly expensive.
The OSM data is licensed in a way that allows redistribution and the project actively encourages people to do this. Clients are allowed to aggressively cache or mirror the data, something which Google or Bing maps do not allow.
Did anyone ever really make a concerted effort to sell tablets to the masses before Apple?
Yes, but the problem was that they were all hardware manufacturers. Apple was the only company that could make the hardware and also rewrite the software stack to be entirely based on touch for input. Earlier tablets ran operating systems designed for either mouse or pen input and most tried to run legacy software. Few people liked them, because the UI sucked. If Apple had shipped the iPad running OS X and existing Mac apps, it would have been a failure.
If you're not failing 90% of the time, then you're probably not working on sufficiently challenging problems
I think I'd find failing 90% of the time completely demoralising, but it's certainly true that if you never fail then you're probably not exploring really interesting possibilities.
Since 1988.
The advantage of the cloud is that it lets you make assumptions that won't be tested until you are unable to restore your backups? Actually, that sounds about right...
I use sshguard, but for a while I was seeing several hundred failed login attempts a day, even though they were blocked after 3 fails (for increasing periods). It may be that after a year my pf table of blocked addresses now includes the entire botnet, or it may be that they just gave up (password logins are disabled on that machine anyway). Or it might just be that the botnet found something more profitable to do than attack a server with half a dozen users...
You're misreading what I said. If you disable the account after n retries, irrespective of where they come from, then that allows a very simple and low bandwidth DoS - just try to log in n times from your botnet and then the real user can't log in at all.
For those interested in a technical / long explanation, see below along with citation links.
Wow. You have way too much time in your hands to come up with such a detailed list. Thanks! I hadn't realised that baud rates kept increasing after V.23bis - I remember reading at the time claims that 2400 baud was the maximum possible with a phone line and that further advances would have to come from other things. The Trellis modulation paper looks like a good way of killing a chunk of boring time on a train next week...
Invalid definition for this conversation
No, it's entirely relevant. The entire point of this story is about people leaving broadcast TV for other ways of getting the same content. It's also the legal definition in the country where I live.
By your definition, every analog TV in the US stopped being TVs when we switched to digital
Yes, if they didn't have a digital decoder box attached, that's correct. They're just analogue monitors then. If they have a set top box that can decode ATSC attached, however, then they are still TVs, because they can receive broadcast television.
The problem is that botnets have a lot of IP addresses. They can do one try from one machine then another from the next. If you disable the account entirely after a certain number of failed logins, you've just created a simple DoS attack. If you disable it just from that IP, it doesn't matter because it will just try from another. There are some realtime block lists that you can use to reject things, but these add another attack route that can let someone who can spoof DNS prevent you from logging in to your own machine...
I think you're underestimating the subtlety of the intelligence agencies. Blackmail works better than threats.
The mockery was not directed at Palin for being able to see Russia or not being able to see Russia, it was directed at Palin for claiming that being able to see the tip of Russia, about as far away from Moscow as New York is from Paris, had anything to do with her competence with regard to foreign policy.
A TV is a device that receives broadcast television. I don't have a TV. I therefore don't need a TV license.
I've been using DuckDuckGo as my search engine for over a year now. It uses (among other things) Bing via Yahoo's BYOSS API, but it doesn't pass any information about me to them and it doesn't use tracking cookies and works via SSL by default. The search results are usually good enough, and the few times they haven't been I've tried Google and got equally bad results there. The only Google service that I do regularly use is YouTube (which ClickToPlugin makes vaguely useable), and that's hardly something I couldn't live without.
IT's like listening to old people be angry because their VCR flashes 12.
Why not be angry about that? Adding a small rechargeable battery that keeps the clock powered when the mains goes out would add, maybe, 50 cents to the cost of production. Not doing it is just indicative of poor design and lack of attention to detail.
A large part of good UI design is having sane defaults. Saying 'no, it doesn't suck because you can log in and then click on the don't suck checkbox and then it's fine' does not actually make it a good UI.
Because, historically, closed communication systems, once they reach a certain critical mass, have only been displaced by more open systems. A new player can't compete easily an entrenched player because of network effects, but lots of new players can gain control of niches that the major player doesn't satisfy and if they all interoperate then between them they can become as large as the major player - then network effects work the other way.
Google realised this with Google Talk, which is a federated XMPP deployment. On its launch day, Google Talk users could talk with millions of existing XMPP users. The XMPP installed base was probably smaller than AIM or MSNM, but it was already fairly large.
If Google pushed an open standard (hopefully not Diaspora, but something actually designed via a process involving actual thought) then they'd get interoperability with the group of people who don't trust companies like Facebook and Google - probably by now the largest set of people who aren't already on Facebook - and with anyone else who wanted a small slice of this market. Instead, they pushed something that has more or less the same set of disadvantages as Facebook, but without the one real advantage that Facebook has: lots of existing users.