I got given a QI charger a little while ago... ended up loving it so much i bought a couple more and it has a number of advantages. Firstly for my phones i always end up with a fairly rugged case which makes it hard to find a dock that works, but even when I do find a workable dock its only real advantage is that it keeps the phone upright but its still "fiddly" (which gets slowly worse over time as the connector starts getting worn out)... Its rather hard to explain why the charger has been something i've come to love without videos/photos etc, but everyone i know who has one that works loves it...
Note the use of the term "works" here... a couple of people i've know who've had them hated them because they simply weren't a very good charger (took too long to charge), ironically the ones i ended up with are some of the cheapest but in terms of phone charging they are impressive (what used to take about 30 minutes now takes perhaps 40 as an example). Which leads me to say one last thing - they may seem almost irrelavent in terms of convenience but they are actually alot more convenient then they may first seem to be and they are also now so cheap they're easy to get a hold of.
just to be clear - i dont make, sell, hold patents, shares or make any money directly or indirectly from any form of wireless charging (afaik)...
A HUD is something i've wanted since... so long ago i cant even remember when it started, but it was certainly around the time of me owning an amiga 500 and really dont understand the hate factor behind google glass nor where its coming from...
But then, i dont entirely get the bluetooth headset hate either, nor why some people find people talking on mobile phones in public to be a nuisance.
To me, i chalk it up to a single simple thing - hatred of technology and im exactly the opposite of that
The *ONLY* thing i can understand about what might make people dislike the idea of google glass is the camera, thats a feature i can understand people not being happy with.
But then, being able to record people in secret has been a simple thing for quite a long time now and you can do it for not even a fraction of the cost of what google glass is. In reality, at least when i see someone is wearing google glass i know they have a camera pointed at me, but if you've seen a watch, pen, button or any other form of hidden video/audio recording device (available these days for under ~$50) then google glass holds very little to threaten any reasonably intelligent person.
First, google has said ad's are fobidden on glass... secondly, "well over a $1000" was for the dev version, the production version is rumoured to weigh in around ~$500 (rumoured being the appropriate word)...
But, if you bothered reading, both these things have been made abundantly clear and if you knew even a single thing about host most dev type stuff happens with hardware like this, the initial dev releases are almost always expensive compared to their later production cousins... but then, again, this is common knowledge to anyone who's been alive more then 15 years (have u even heard of economies of scale or simple things such as this?)... I honestly dont get how such a stupid post gets "3, insightful", but then again, welcome to slashdot right?
Yeah, im being deliberately insulting, but if your going to make such a retarded, fud-based and false post, you deserve no less...
I say yay because i really dont like sip, i think its a horrible protocol and anything that would end its life is fantastic (even if it is a patent troll). Though its hard to see if BT are claiming ownership of any tech that does voip to pstn or just SIP.
With any luck, someone will develop a useful protocol to replace it, though my hopes arent high.
Say what you guys will, personally i've been following EOMA-68 for a while and for a good reason. Ultimately a long time ago, i've been wanting to do what calxeda now do (around the time the first nslu2's came out) and was really excited when calxeda started talking the talk... then they released a box that is way over engineered and way too expensive for the role (not to mention rather inflexible).
With eoma-68 you could do the same thing, only the price would be far more suitable and the server boards inside the chassis "standardised" on a simple interface anyone can implement. The backplane for putting 100 of these boxes into a 2ru chassis is quite simple also. Think: a replacement for x86 based virtualisation and vps's
So yes, you may look at it and see a slashvirt if you will, however there are those of us in tech that see a real purpose to it also..
Well, the mk802 is a system on a stick (which isnt what im refering to)... but for the same price you can get an mk802, you should be able to get its dev-board cousins - i.e. what the rpi is (which typically will have all the connectors on the arm chip exposed, gpio included) - and those dev boards have been around as long as the mk802 has.
As I said in the post above though, i love how the rpi has gotten this ground swell of support following it, and its spawned some very kewl things (all of which work equally well on the a10 hackberry/cubieboard/etc type boards), but my point is more that bang for buck, the rpi is actually quite expensive... but consider, for me the rpi is 53$ (shipping, etc) and an a10 based board i can get at ~$75 (inc shipping + cables) and for that i get dual core, 1.2ghz and 1gb of ram - this is not a small difference (and if it were being sold in the quantities that the rpi is, it'd be cheaper still).
I guess one thing that does irritate me about it all is that it starts out as a $25 board (roughly?) and then between me and the people who make them there are 3 layers of idiots just reaping money out of the whole thing - i have had to deal with both rs and element14 for quite a long time (or the companies which they were originally) and they really couldn't have a chosen a worse set of distributors in Australia for something like the rpi.
if i login right now and go thru their order page, its $53 inc shipping to Sydney...
As i was saying, for 60$ you can get an a10 based board (dual-core, 1.2-1.5Ghz + 1gb of ram) shipped (at 70-75, you'll get every cable you'll ever need too) - and it has every connector the rpi has. But there is a world of difference between 256 and 512 of ram, and another world of diff between 512 and 1024...
Dont get me wrong, the rpi has done one wonderous thing - its gotten people very firmly interested in a architecture i love (and have since around the time of the nslu2 - which in reality was not really an arm board, but based off the arm architecture). the things (and projects) it has spawned are fantastic...
Its just that when i sit there and compare an rpi to those a10 boards, i just dont get why people are all going "yes, for 10-20 more, i could get a board with 4 times the grunt, but i'd rather save the 20$"... sure in a third world, a $20 diff is important, but to the very vast majority of people who are doing projects with them, they are ultimately an expensive board for what they do and should probably be a little cheaper.
Personally, i've never been interested in an rpi, im really very into arm based tech, but at least in australia where i am, the rpi ends up being quite expensive. Its been nice as a project for the general community to work on cause it seems to have focused everyone on a single arm board and so alot of projects have errupted from it. However as a piece of hardware, by the time you could get it in australia there were (and still are) much better options. Ultimately you'll end up spending $50 here, and then theres the stupid decision about the distributors which charge rediculously for shipping over here.
Ultimately though it was the specs on the board that switched me off in the end (which now with the larger one, still are a little light imho). For around $60 you can pick up several a10 based options which have much better cpu's and memory - often when you factor in shipping, they end up being cheaper. But then, im not exactly a third-world person, so the cost of the board was never really that important anyway. i.e. at the time the hackberry was $75 for a dual-core 1.2GHz and 1g of RAM, vs the RPI's ~$30 for 800mhz and 256M. You can either look at that as 3times the price or (as i did) a fairly insignificant price difference for a much better board and they're still cheap enough for me to own a number of them.
Lets ignore the fact that its 3d for a moment and concentrate on your main points... Is it novel? yes i personally thing so, but novel is in the eyes of the beholder. Is it awesome? absolutely from a developer point of view... is it cross platform - yes it is. Chrome and firefox can both be downloaded for every platform that oracle java can be (and probably built for many more too).
Then theres the version dependency fun. Consider i wrote a RTF document editor as java applet, im faced with the scenario of 3 very annoying dependencies i have to code for - first the browser version, next the java version and lastly the OS itself. The browser and version of it in that case can be problematic in some cases, but largely they just fork off to java, and this is where the real fun begins because the current version of java out there still rotate around 1.6 and 1.7 and sadly I still have 1.6 because apps written for 1.6 often have problems in 1.7... Next lets just consider windows and macos and the fun (which still exists) of making a java app run on both systems - throw linux into the mix (and i do) and it gets far worse... These days, your chances of the end user actually having java or flash or some other add-on to support your app aren't always the best odd's either.
But now the browser itself is becoming capable enough of replacing that middlewear dependency. Consider simply this, how many websites do most people go to today that have java applets on them? the answer is so very very small its really not funny. Take a look at google docs for example, how much of that would have to be written in some other addon-required code (such as java) if web browsers didnt become fundamentally more capable at doing things which java does?
When you add something like 3d into that scenario, your pain levels go up orders of magnitude. How many 3d java app's are there out there? well, theres minecraft and.... 3d in java is painful when you start trying to do something even remotely complex, minecraft is by design very simple in its 3d implementation, it barely uses more then the primitives and so its relatively easy to make work almost anywhere, but you wouldn't have to add many things to minecraft for the 3d cracks to start to appear (shaders for example).
Lastly, take a look at the code that was written to make the 3d stuff in the article work - if you can do the same with a java applet anywhere near as simply as this has been done, i'd be very very surprised.
as everyone's stated, what you've done so far is correct.. IMHO, controllers are well worth the money - though shop around, cause (again, IMHO) juniper and cisco are way too expensive for what they are.
What a controller will give you is a unified simple way of managing it all. I.e. configure it in one spot rather then every AP. They also often include things like portals, authentication services and firewalls. I.e. a central CA for using certificate based auth, a captive wifi portal for open access points that go to the internet or stuff like that.
Where that becomes GREAT is trying to debug stuff, when you get past 4 AP's it starts to get a little tedious making sure every AP is configured correctly (i.e. same SSID, same authentication info), and gets really hard to maintain channel separation effectively. Alot of controller based systems will distribute the channels well based on the topology of your AP network, and that is very handy.
All of this is doable manually, what a controller can do that you cant do anywhere else is force handoff from one AP to another. AP Clients typically head for the closest AP based on signal strength alone and that can get a little annoying because you'll often end up with several AP's that are flooded and others that are barely used, controllers can manage that and push clients off one AP and tell them to use a different one.
The other bit that is mighty hard to do with out a controller is running multiple SSID's from the same AP's connected to different networks (and often the firewall in the controller plays a part in this too). It can be handy in some situations to have a "visitor" SSID thats open access but only gets internet along side an "internal" SSID that gets on your internal network and maybe authenticated via certificates. Controllers handle that very well.
I work with Juniper and Cisco on no-wifi, and for the last 18 months we've been doing wifi with aruba too... couldn't agree more, they're quite fantastic and gained something of a reputation for doing wifi well (well deserved IMHO).
This is *NOT* what WDS was designed to do. There seems to be quite a lot of people under the impression that if you want multiple access points co-operating with one another such that clients can roam between them seemlessly, you need WDS. Not sure where that came from but its got nothing to do with that.
WDS is about peer-to-peer AP connections such the data is travelling wirelessly between access points, and while WDS can be the "backbone" of a seemlessly-roaming SSID-consistent WiFi network, its an inherently flawed system. This is typically used for places where you need to bridge networks wirelessly when you cant put down a cable (for eg, you might have two offices across the road from one another).
WDS will also chew up a considerable amount of wifi bandwidth doing this (and the problem gets exponentially worse as you add more AP's/clients).
The point being though that WDS wasnt designed for the purposes of providing distributed access to a wifi network with a single SSID, but to allow AP's to also be clients to each other while still being AP's.
Ultimately the way the guy describes his setup is the correct method of deployment, multiple AP's with the same SSID and encryption parameters, thats all there is to it.
i had the same thought... and yes i really do hate the lack of sd card on nexus devices (though for other reasons)... however, for media a better choice is actually a usb otg adaptor - you can generally get faster, cheaper and larger storage on the end of a usb stick then you will in a microsd card and while a bit larger, they're not inconveniently so (imho)...
Im loving my nexus 10, i really wasnt expecting to cause the reviews have been iffy at best, but it really is (to me) exactly what i've wanted in a tablet for a long time.
But like you, i do despise the nexus range for its "no-sd-card" policy and its been (so far) my only real gripe with the entire nexus lineup.
NAT is and never was a stupid hack, people make the assumption it stemmed from the exhaustion of ipv4 , but that isnt true... as many people do, you've made the same invalid assumption as everyone else - and thats not meant to be an insult, just a fact of life. The reality is is that we've been working with nat so long that it really doesnt break much any more, and that which it does is work-around-able. Personally I come form the corporate enterprise world and nat is and always will be a reality there (for many reasons)... When you consider the various grades of people that use the internet (home users, geeks, corporates and ISPs, etc), everyone of them has no reason to fear nat, and ipv6 wont change anything about that... IPv6 wasnt designed with any notion around "lets make something that gets rid of nat", not even close. It certainly was intended to make the address space larger, but it wasnt aimed at nat.
The only diff between me (as a network engineer) and many other enterprise grade guys like me is that im dumb enough to say "nat is important, it has a purpose and i do like it" where most aren't dumb enough to say that on slashdot where the various puritan opinions will inevitably turn me into a troll.... which im really not meant to be...
But the fact remains - nat has a purpose, a valid purpose and it will be here forever.
Consider though - just because something is abundant, doesnt mean its necessarily available to you. By that I mean if your ISP decides to only give you one dynamic ipv6/64 subnet, what then? most home consumers wont notice, but a networking nerd like me? well... there are lots of arguments you could enter into here like "find a new isp" - none of which are worth having. But a brief tour of the nasty things ISP's (not all) have tried to do to make a couple of extra dollar's would probably give a few hints as to why ipv6 nat (and its implementation in linux) is very very important.
You could also go and get a CCIE or JNCIE and start working in the corporate enterprise juganaut world for a decade and i guarantee when get to the end of that you'll change your tune....
IMHO, this is one of the more important features added in 3.7. People will hate me for saying it, but I personally love NAT and was really disappointed when initially the line was "we will never do ipv6 nat in the linux netfilter components".
Its not the pay to play model that is bothersome, this is nothing new to MS (or many companies)... MS are experts at licensing schemes that ensure they rip as much cash out of people as possible, and they've gotten better at that over the years... i.e. licensing to the nth degree... for example, you used to buy a copy of exchange, then you bought a license for exchange, and these days you buy a license for so many different parts of exchange including per-client licensing...
Thats all nothing new and the pay to play model is noting new either. Though, the idea that people will only get software from a MS controlled central location (such as what occurs in the smart-phone application market model) would certainly take pay to play to a new place MS havent been before (in that they'd be tearing money out of developers of MS-targetted software in a direct, "get a percentage of their profits" way).
If anything, i'd say the media industry learnt from software licensing, not the other way around...
My reference to the MS strategy is simply "bring out something new every year and stop trusting your users"... i.e. walled garden and new varnish on old technology.
Its a shame that MS seem to feel they have only one option now... "be like apple".
Half the reason they thrive so well in corporate-enterprise-juganaut land is simply because they aren't apple and dont behave like them. A release every year is going to be an utter nightmare for a decent sized enterprise, but i guess it depends on what "next version" really means. Is it going to just an incremental update similar to what service packs used to be? In which case, the actual OS update will probably less painful, but there will be pain to be had in other places (namely licensing).
I really wouldn't be cheering for this idea if i were in a corporate desktop support role, thats for absolute certain.
Even given the job that i do (which falls into the systems integrator role), it doesn't sound good... whats it going to mean for certification? oh the pain.... then that comes with its own set of licensing nightmares (the SI role).
Still, as a linux-lover, i can only say "i love where apple and MS are taking their OS's because they seem to be working very hard to make linux as attractive as possible".
I would rule out the 360 for the simple reason that the difference between the ps3 and 360 for what you want is very slim, your not going to get many games on a 360 that your 9yr old will enjoy that arent also already available on the ps3. Ultimately you'll have another console, but it'll just be different, not unique really. Kinect can be kinda fun, but its a limited experience in many ways.
If you want to buy something, the Wii is the best choice. You'll find alot of good games that will appeal to a 9yr old. Games the whole family can get in on and are only available on the Wii. Its a very family focused entertainment console and you'll get quite a number of hours of enjoyment out of it. On top of all this, the Wii is cheap and second hand games for it are (not yet) hobbled by scum-sucking games developers who seem to think they have a right to make money on a second hand game.
The Wii U looks ok, but its hard to really make a knowledgeable choice about it just because of the limited info available for it.
Im trying hard to stick to the question and not say things like "you could get another console, or...."
Its not necessarily about an advantage of either really.. clang/llvm is really more about modularity and focusing on a more flexible toolchain.
As an example, if you want to compile from c-> binary in linux, theres quite a command path depth to it (i.e. multiple processes take place in serial to eventually produce a binary). The problem with gcc is that the code has become messy and its modular approach doesnt really help when it comes to adding a new language or a new target (i.e. F# perhaps, or ARM as a target). It also carries alot of legacy cruft with it that has been there since the dawn of time.
Ultimately when you do want to add a new language or target, you almost end up re-writing the entire gcc stack which is quite painful.
clang/llvm is really about getting rid of that pain with a more modular approach so if you want to add a language or a target, you just focus on your goal not the entire toolchain. As a Plus, its also gotten rid of some of the antique cruft which is at least part of the reason its often faster. IMHO - its faster in some situations, but not all.
None of this really matters much to the end user (except perhaps the speed bit)... What does matter (and what you will care about) is as clang/llvm matures you'll find developers more likely to target clang/llvm for compiler toolchains with new languages or targets.
The other thing you might care about as well is that its better with errors, they are alot clearer and more useful and for that see http://clang.llvm.org/diagnostics.html however, that only extends to clang really.
I personally hate and despise people who put non-rackmount kit in racks...
We use various devices.. mostly all 1ru servers of various configs... for eg there are a couple of mini-itx 1ru servers we have that have e350 based mini-itx boards (i really love the e350/e450 boards)... not quite as cheap as the hp n40 microserver, but at least its a rack format.
Then we have a few that run virtualisation here and there for some tasks using kvm (some of those too have e350's in them as the e350's do have the virt'n extensions unlike the intel atoms)... we also have a few that run intel based i3/i5/i7 mini-itx boards... they're quite nice when you need some extra grunt...
some others are based on super micro boards as well though (which are quite cheap and run core i3/i5/i7 cpus rather the xeons). Then some others are old 1ru xeons we no longer need for server tasks...
Oh, two other things (and i would test it if i could, but my better half has stolen it). It doesnt require special paper and im pretty sure when the tracking device is plugged in it comes up as a mouse and operates as such (in relative mode)... but dont quote me on that.
I have a CHEAP digital pen (cost me 25$) called the "greenpoint mobile notetaker" (which i think is a pegasus notetaker rebadged).... its ultra simple. it works under linux and what the linux software gives you is a simple svg map of what you drew on a piece of paper. Its just a normal pen with a little tracking unit that somehow tracks everything you write... I dont use it too much, but the times i have its not failed me so far.
I find it works ok, but i've not really used any other digital pens, so i have no point of comparison - but at 25$ (which was on sale at the time) i just went "sure why not" and later found out it supported linux (which was a nice surprise).
i've seen many ssd's die in enterprise SAN kit and will even go so far as to say as a percentage they die far more often then spinning rusty metal.
However, this is because of how they are used in SAN. Often they are used in a multi-tier way, where most frequently accessed data is pushed up to the SSD's to allow quick access, so they get hit the hardest.
I would be guessing your asking this question simply cause its easier to understand why a part-in-motion can slowly die over time where something thats in silicon shouldnt. You'd probably be surprised to know that alot of drives die a controller death, not a platter/motor death. Typically a platter/motor death usually means a badly made drive (mostly because makers of spinning rusty metal have gotten very good at the mechanics behind them) and while the mechanics in a drive will slowly wear over time, typically something silicon in the controller goes first. The exception to that is where drivers are physically in motion not of their own (i.e. laptops for example), often then the drive shaft in the drive itself starts to get wear unevenly and that usually gets worse over time (or at least, this is what i've been told).
Some SAN makers will even put hdd's through their paces first to make sure they actually perform ok - for eg, they'll measure vibration, etc (i.e. the mechanical components) to make sure a drive is up to spec before it goes into their kit - they cant really do the same as easily with silicon as theres not much to work off that can be measured - so often in enterprise grade SAN's, silicon dies before mechanical.
Having said all that, i honestly cant wait for the death of spinning rusty metal for the simple reason that ssds should (and havent yet) taken on forms which would be much more useful - why use a sata 2.5" format when you could have much better geometries for example? Then theres the interfaces we use where were really designed with hard drives in mind... but thats an entirely different issue.
As for "if its all the same, why doesnt it die at the same time?"... because at a fundamental level, it isnt the same. When we make anything no matter what it is it, the materials go thru some form of refining process where impurities are removed. Its impractical and near impossible to get anything 100% pure (not to mention entirely uneconomical - you might pay $1/tonne for 90% pure iron and $100/tonne for 91% pure iron as an example). The nature of where we get those materials often means the impurities vary greatly in composition in small spaces of time, hence why two hdd's sitting next to each other on the assembly line might mean one will die after 2 days and the other will die after 200 years. Theres also other components to this in that if you looked at iron (again an example, but true of most metals) under a microscope you'd find that it isnt a uniform substance, its quite grainy so to speak and those grains vary considerably. This too impacts the nature of the substance and its longevity. Using iron as an example, when its cooled its very hard to get a truly consistent cooling of it across the entire piece and the cooling determines how those grains form - i.e. grains in the center of the material will be quite different to grains at the edge and so forth. Thats just a small number of things that explain why consistency isnt quite as 100% as it might appear to be on the surface, there are quite alot of factors that come together to effect materials we use. Ultimately the way we choose the materials we produce things with is by tolerance, i.e. i expect 99% of my metal you sell me to fall into 90% pure and have x tensile strength or any number of variables you might consider important to your manufacturing process but even then its never 100%, you always except at some point that you'll get raw materials that'll fall outside those tolerances and as with everything on the planet its a trade off between price vs quality!
At first i was curious why, but when it comes to the insurance side of things its makes alot more sense...
However, how long until this starts impacting everything you do... obviously in this case things that kill you out-right arent a problem for insurance companies (i would assume). But consider the possible outcomes of such a thing on the following activities: - scuba diving - riding a motorbike - sky diving - driving a car - using a 3d printer (printing in abs for example produces bad fumes, depending on what you want to believe)
They are just random examples but all of these have a potential to mess you up in some nasty ways, and in some cases your more likely to end up damaged by them then by smoking (i know a few people who sky dive and not one of them hasnt had some injury thats put them out of action for a reasonable period of time). So before you cheer what appears to be a win for the non-smoking consortium, consider the potential damage to your own after-hours hobbies. i.e. anything an insurance company can say is a "risk" is a potential "sorry, we cant employ you" and that is a rather worrying outcome.
the general role of any company these days seems to be to be able to hide from any form of moral behaviour... the larger the company the more likely they are to behave "within the law" but despicably so. Essentially running a company means you can abandon any form of ethical or moral behaviour - and apparently that's fine.
Patents are a good example of that. They should foster innovation, but they dont for so many reasons. Take apple for example "slide to unlock", "searching the phone AND the internet" these are not innovation, they are simple ideas anyone could have come up with from 5 minutes in front of a phone and yet apple patent them because they can and it gives them leverage to abuse everyone else. In just about any other part of society that kind of behaviour would be considered highly corrupt, yet when it comes to company ethics well there are none and so things like this happen. The criteria for "invention" have become so unbelievable low that simply 3 lines of code might be considered invention if you spin it the right way - for a company this is acceptable, but not from an ethical perspective.
The reality is the patent system should award innovation, but all it achieves is a platform for abuse. From that point alone it probably deserves to die. The problem is that it really wont change much, will apple (and others) suffer terrible losses (which imho, they deserve) from the loss of the patent system? no, probably would work in their favour. Sure they do have patents, but without patents they wouldnt have to pay smaller people for the inventions they license (assuming they do). It might mean people like samsung could take what they want from anyone, copy it, the mass-market it more effectively and just destroy people. That too is wrong.
Sadly, if I working alone came up with some awesome bit whatever that really was "innovation" and "invention" i'd probably end up abused for it. Some large company would simply realise that as a singular individual i probably wouldnt be able to defend myself from their battery of lawyers and so i'd get destroyed anyway. So with or without the patent system, your kinda screwed anyway.
But then having said all that, isnt this exactly what the american capitalist ideal is all about anyway? So perhaps from an american perspective, this is exactly what the patent system was setup to achieve and its doing a great job. I personally just cant see how people can look at the things apple, microsoft, and many others have patented (and used against people) and go "yeah, thats innovation and invention". Its kind of pathetic really.
Slashdot Mirror
I got given a QI charger a little while ago... ended up loving it so much i bought a couple more and it has a number of advantages. Firstly for my phones i always end up with a fairly rugged case which makes it hard to find a dock that works, but even when I do find a workable dock its only real advantage is that it keeps the phone upright but its still "fiddly" (which gets slowly worse over time as the connector starts getting worn out)... Its rather hard to explain why the charger has been something i've come to love without videos/photos etc, but everyone i know who has one that works loves it...
Note the use of the term "works" here... a couple of people i've know who've had them hated them because they simply weren't a very good charger (took too long to charge), ironically the ones i ended up with are some of the cheapest but in terms of phone charging they are impressive (what used to take about 30 minutes now takes perhaps 40 as an example). Which leads me to say one last thing - they may seem almost irrelavent in terms of convenience but they are actually alot more convenient then they may first seem to be and they are also now so cheap they're easy to get a hold of.
just to be clear - i dont make, sell, hold patents, shares or make any money directly or indirectly from any form of wireless charging (afaik)...
A HUD is something i've wanted since ... so long ago i cant even remember when it started, but it was certainly around the time of me owning an amiga 500 and really dont understand the hate factor behind google glass nor where its coming from...
But then, i dont entirely get the bluetooth headset hate either, nor why some people find people talking on mobile phones in public to be a nuisance.
To me, i chalk it up to a single simple thing - hatred of technology and im exactly the opposite of that
The *ONLY* thing i can understand about what might make people dislike the idea of google glass is the camera, thats a feature i can understand people not being happy with.
But then, being able to record people in secret has been a simple thing for quite a long time now and you can do it for not even a fraction of the cost of what google glass is. In reality, at least when i see someone is wearing google glass i know they have a camera pointed at me, but if you've seen a watch, pen, button or any other form of hidden video/audio recording device (available these days for under ~$50) then google glass holds very little to threaten any reasonably intelligent person.
wow, talk about mis-informed...
First, google has said ad's are fobidden on glass... secondly, "well over a $1000" was for the dev version, the production version is rumoured to weigh in around ~$500 (rumoured being the appropriate word)...
But, if you bothered reading, both these things have been made abundantly clear and if you knew even a single thing about host most dev type stuff happens with hardware like this, the initial dev releases are almost always expensive compared to their later production cousins... but then, again, this is common knowledge to anyone who's been alive more then 15 years (have u even heard of economies of scale or simple things such as this?)... I honestly dont get how such a stupid post gets "3, insightful", but then again, welcome to slashdot right?
Yeah, im being deliberately insulting, but if your going to make such a retarded, fud-based and false post, you deserve no less...
I say yay because i really dont like sip, i think its a horrible protocol and anything that would end its life is fantastic (even if it is a patent troll). Though its hard to see if BT are claiming ownership of any tech that does voip to pstn or just SIP.
With any luck, someone will develop a useful protocol to replace it, though my hopes arent high.
Say what you guys will, personally i've been following EOMA-68 for a while and for a good reason. Ultimately a long time ago, i've been wanting to do what calxeda now do (around the time the first nslu2's came out) and was really excited when calxeda started talking the talk... then they released a box that is way over engineered and way too expensive for the role (not to mention rather inflexible).
With eoma-68 you could do the same thing, only the price would be far more suitable and the server boards inside the chassis "standardised" on a simple interface anyone can implement. The backplane for putting 100 of these boxes into a 2ru chassis is quite simple also. Think: a replacement for x86 based virtualisation and vps's
So yes, you may look at it and see a slashvirt if you will, however there are those of us in tech that see a real purpose to it also..
Well, the mk802 is a system on a stick (which isnt what im refering to)... but for the same price you can get an mk802, you should be able to get its dev-board cousins - i.e. what the rpi is (which typically will have all the connectors on the arm chip exposed, gpio included) - and those dev boards have been around as long as the mk802 has.
As I said in the post above though, i love how the rpi has gotten this ground swell of support following it, and its spawned some very kewl things (all of which work equally well on the a10 hackberry/cubieboard/etc type boards), but my point is more that bang for buck, the rpi is actually quite expensive... but consider, for me the rpi is 53$ (shipping, etc) and an a10 based board i can get at ~$75 (inc shipping + cables) and for that i get dual core, 1.2ghz and 1gb of ram - this is not a small difference (and if it were being sold in the quantities that the rpi is, it'd be cheaper still).
I guess one thing that does irritate me about it all is that it starts out as a $25 board (roughly?) and then between me and the people who make them there are 3 layers of idiots just reaping money out of the whole thing - i have had to deal with both rs and element14 for quite a long time (or the companies which they were originally) and they really couldn't have a chosen a worse set of distributors in Australia for something like the rpi.
if i login right now and go thru their order page, its $53 inc shipping to Sydney...
As i was saying, for 60$ you can get an a10 based board (dual-core, 1.2-1.5Ghz + 1gb of ram) shipped (at 70-75, you'll get every cable you'll ever need too) - and it has every connector the rpi has. But there is a world of difference between 256 and 512 of ram, and another world of diff between 512 and 1024...
Dont get me wrong, the rpi has done one wonderous thing - its gotten people very firmly interested in a architecture i love (and have since around the time of the nslu2 - which in reality was not really an arm board, but based off the arm architecture). the things (and projects) it has spawned are fantastic...
Its just that when i sit there and compare an rpi to those a10 boards, i just dont get why people are all going "yes, for 10-20 more, i could get a board with 4 times the grunt, but i'd rather save the 20$"... sure in a third world, a $20 diff is important, but to the very vast majority of people who are doing projects with them, they are ultimately an expensive board for what they do and should probably be a little cheaper.
Personally, i've never been interested in an rpi, im really very into arm based tech, but at least in australia where i am, the rpi ends up being quite expensive. Its been nice as a project for the general community to work on cause it seems to have focused everyone on a single arm board and so alot of projects have errupted from it. However as a piece of hardware, by the time you could get it in australia there were (and still are) much better options. Ultimately you'll end up spending $50 here, and then theres the stupid decision about the distributors which charge rediculously for shipping over here.
Ultimately though it was the specs on the board that switched me off in the end (which now with the larger one, still are a little light imho). For around $60 you can pick up several a10 based options which have much better cpu's and memory - often when you factor in shipping, they end up being cheaper. But then, im not exactly a third-world person, so the cost of the board was never really that important anyway. i.e. at the time the hackberry was $75 for a dual-core 1.2GHz and 1g of RAM, vs the RPI's ~$30 for 800mhz and 256M. You can either look at that as 3times the price or (as i did) a fairly insignificant price difference for a much better board and they're still cheap enough for me to own a number of them.
Lets ignore the fact that its 3d for a moment and concentrate on your main points... Is it novel? yes i personally thing so, but novel is in the eyes of the beholder. Is it awesome? absolutely from a developer point of view... is it cross platform - yes it is. Chrome and firefox can both be downloaded for every platform that oracle java can be (and probably built for many more too).
Then theres the version dependency fun. Consider i wrote a RTF document editor as java applet, im faced with the scenario of 3 very annoying dependencies i have to code for - first the browser version, next the java version and lastly the OS itself. The browser and version of it in that case can be problematic in some cases, but largely they just fork off to java, and this is where the real fun begins because the current version of java out there still rotate around 1.6 and 1.7 and sadly I still have 1.6 because apps written for 1.6 often have problems in 1.7... Next lets just consider windows and macos and the fun (which still exists) of making a java app run on both systems - throw linux into the mix (and i do) and it gets far worse... These days, your chances of the end user actually having java or flash or some other add-on to support your app aren't always the best odd's either.
But now the browser itself is becoming capable enough of replacing that middlewear dependency. Consider simply this, how many websites do most people go to today that have java applets on them? the answer is so very very small its really not funny. Take a look at google docs for example, how much of that would have to be written in some other addon-required code (such as java) if web browsers didnt become fundamentally more capable at doing things which java does?
When you add something like 3d into that scenario, your pain levels go up orders of magnitude. How many 3d java app's are there out there? well, theres minecraft and.... 3d in java is painful when you start trying to do something even remotely complex, minecraft is by design very simple in its 3d implementation, it barely uses more then the primitives and so its relatively easy to make work almost anywhere, but you wouldn't have to add many things to minecraft for the 3d cracks to start to appear (shaders for example).
Lastly, take a look at the code that was written to make the 3d stuff in the article work - if you can do the same with a java applet anywhere near as simply as this has been done, i'd be very very surprised.
as everyone's stated, what you've done so far is correct.. IMHO, controllers are well worth the money - though shop around, cause (again, IMHO) juniper and cisco are way too expensive for what they are.
What a controller will give you is a unified simple way of managing it all. I.e. configure it in one spot rather then every AP. They also often include things like portals, authentication services and firewalls. I.e. a central CA for using certificate based auth, a captive wifi portal for open access points that go to the internet or stuff like that.
Where that becomes GREAT is trying to debug stuff, when you get past 4 AP's it starts to get a little tedious making sure every AP is configured correctly (i.e. same SSID, same authentication info), and gets really hard to maintain channel separation effectively. Alot of controller based systems will distribute the channels well based on the topology of your AP network, and that is very handy.
All of this is doable manually, what a controller can do that you cant do anywhere else is force handoff from one AP to another. AP Clients typically head for the closest AP based on signal strength alone and that can get a little annoying because you'll often end up with several AP's that are flooded and others that are barely used, controllers can manage that and push clients off one AP and tell them to use a different one.
The other bit that is mighty hard to do with out a controller is running multiple SSID's from the same AP's connected to different networks (and often the firewall in the controller plays a part in this too). It can be handy in some situations to have a "visitor" SSID thats open access but only gets internet along side an "internal" SSID that gets on your internal network and maybe authenticated via certificates. Controllers handle that very well.
I work with Juniper and Cisco on no-wifi, and for the last 18 months we've been doing wifi with aruba too... couldn't agree more, they're quite fantastic and gained something of a reputation for doing wifi well (well deserved IMHO).
WRONG!
This is *NOT* what WDS was designed to do. There seems to be quite a lot of people under the impression that if you want multiple access points co-operating with one another such that clients can roam between them seemlessly, you need WDS. Not sure where that came from but its got nothing to do with that.
WDS is about peer-to-peer AP connections such the data is travelling wirelessly between access points, and while WDS can be the "backbone" of a seemlessly-roaming SSID-consistent WiFi network, its an inherently flawed system. This is typically used for places where you need to bridge networks wirelessly when you cant put down a cable (for eg, you might have two offices across the road from one another).
WDS will also chew up a considerable amount of wifi bandwidth doing this (and the problem gets exponentially worse as you add more AP's/clients).
The point being though that WDS wasnt designed for the purposes of providing distributed access to a wifi network with a single SSID, but to allow AP's to also be clients to each other while still being AP's.
Ultimately the way the guy describes his setup is the correct method of deployment, multiple AP's with the same SSID and encryption parameters, thats all there is to it.
i had the same thought... and yes i really do hate the lack of sd card on nexus devices (though for other reasons)... however, for media a better choice is actually a usb otg adaptor - you can generally get faster, cheaper and larger storage on the end of a usb stick then you will in a microsd card and while a bit larger, they're not inconveniently so (imho)...
Im loving my nexus 10, i really wasnt expecting to cause the reviews have been iffy at best, but it really is (to me) exactly what i've wanted in a tablet for a long time.
But like you, i do despise the nexus range for its "no-sd-card" policy and its been (so far) my only real gripe with the entire nexus lineup.
NAT is and never was a stupid hack, people make the assumption it stemmed from the exhaustion of ipv4 , but that isnt true... as many people do, you've made the same invalid assumption as everyone else - and thats not meant to be an insult, just a fact of life. The reality is is that we've been working with nat so long that it really doesnt break much any more, and that which it does is work-around-able. Personally I come form the corporate enterprise world and nat is and always will be a reality there (for many reasons)... When you consider the various grades of people that use the internet (home users, geeks, corporates and ISPs, etc), everyone of them has no reason to fear nat, and ipv6 wont change anything about that... IPv6 wasnt designed with any notion around "lets make something that gets rid of nat", not even close. It certainly was intended to make the address space larger, but it wasnt aimed at nat.
The only diff between me (as a network engineer) and many other enterprise grade guys like me is that im dumb enough to say "nat is important, it has a purpose and i do like it" where most aren't dumb enough to say that on slashdot where the various puritan opinions will inevitably turn me into a troll.... which im really not meant to be...
But the fact remains - nat has a purpose, a valid purpose and it will be here forever.
Consider though - just because something is abundant, doesnt mean its necessarily available to you. By that I mean if your ISP decides to only give you one dynamic ipv6 /64 subnet, what then? most home consumers wont notice, but a networking nerd like me? well... there are lots of arguments you could enter into here like "find a new isp" - none of which are worth having. But a brief tour of the nasty things ISP's (not all) have tried to do to make a couple of extra dollar's would probably give a few hints as to why ipv6 nat (and its implementation in linux) is very very important.
You could also go and get a CCIE or JNCIE and start working in the corporate enterprise juganaut world for a decade and i guarantee when get to the end of that you'll change your tune....
IMHO, this is one of the more important features added in 3.7. People will hate me for saying it, but I personally love NAT and was really disappointed when initially the line was "we will never do ipv6 nat in the linux netfilter components".
But, thank god someone came to their senses.
Its not the pay to play model that is bothersome, this is nothing new to MS (or many companies)... MS are experts at licensing schemes that ensure they rip as much cash out of people as possible, and they've gotten better at that over the years... i.e. licensing to the nth degree... for example, you used to buy a copy of exchange, then you bought a license for exchange, and these days you buy a license for so many different parts of exchange including per-client licensing...
Thats all nothing new and the pay to play model is noting new either. Though, the idea that people will only get software from a MS controlled central location (such as what occurs in the smart-phone application market model) would certainly take pay to play to a new place MS havent been before (in that they'd be tearing money out of developers of MS-targetted software in a direct, "get a percentage of their profits" way).
If anything, i'd say the media industry learnt from software licensing, not the other way around...
My reference to the MS strategy is simply "bring out something new every year and stop trusting your users"... i.e. walled garden and new varnish on old technology.
Its a shame that MS seem to feel they have only one option now... "be like apple".
Half the reason they thrive so well in corporate-enterprise-juganaut land is simply because they aren't apple and dont behave like them. A release every year is going to be an utter nightmare for a decent sized enterprise, but i guess it depends on what "next version" really means. Is it going to just an incremental update similar to what service packs used to be? In which case, the actual OS update will probably less painful, but there will be pain to be had in other places (namely licensing).
I really wouldn't be cheering for this idea if i were in a corporate desktop support role, thats for absolute certain.
Even given the job that i do (which falls into the systems integrator role), it doesn't sound good... whats it going to mean for certification? oh the pain.... then that comes with its own set of licensing nightmares (the SI role).
Still, as a linux-lover, i can only say "i love where apple and MS are taking their OS's because they seem to be working very hard to make linux as attractive as possible".
I would rule out the 360 for the simple reason that the difference between the ps3 and 360 for what you want is very slim, your not going to get many games on a 360 that your 9yr old will enjoy that arent also already available on the ps3. Ultimately you'll have another console, but it'll just be different, not unique really. Kinect can be kinda fun, but its a limited experience in many ways.
If you want to buy something, the Wii is the best choice. You'll find alot of good games that will appeal to a 9yr old. Games the whole family can get in on and are only available on the Wii. Its a very family focused entertainment console and you'll get quite a number of hours of enjoyment out of it. On top of all this, the Wii is cheap and second hand games for it are (not yet) hobbled by scum-sucking games developers who seem to think they have a right to make money on a second hand game.
The Wii U looks ok, but its hard to really make a knowledgeable choice about it just because of the limited info available for it.
Im trying hard to stick to the question and not say things like "you could get another console, or...."
Its not necessarily about an advantage of either really.. clang/llvm is really more about modularity and focusing on a more flexible toolchain.
As an example, if you want to compile from c-> binary in linux, theres quite a command path depth to it (i.e. multiple processes take place in serial to eventually produce a binary). The problem with gcc is that the code has become messy and its modular approach doesnt really help when it comes to adding a new language or a new target (i.e. F# perhaps, or ARM as a target). It also carries alot of legacy cruft with it that has been there since the dawn of time.
Ultimately when you do want to add a new language or target, you almost end up re-writing the entire gcc stack which is quite painful.
clang/llvm is really about getting rid of that pain with a more modular approach so if you want to add a language or a target, you just focus on your goal not the entire toolchain. As a Plus, its also gotten rid of some of the antique cruft which is at least part of the reason its often faster. IMHO - its faster in some situations, but not all.
None of this really matters much to the end user (except perhaps the speed bit)... What does matter (and what you will care about) is as clang/llvm matures you'll find developers more likely to target clang/llvm for compiler toolchains with new languages or targets.
The other thing you might care about as well is that its better with errors, they are alot clearer and more useful and for that see http://clang.llvm.org/diagnostics.html however, that only extends to clang really.
I personally hate and despise people who put non-rackmount kit in racks...
We use various devices.. mostly all 1ru servers of various configs... for eg there are a couple of mini-itx 1ru servers we have that have e350 based mini-itx boards (i really love the e350/e450 boards)... not quite as cheap as the hp n40 microserver, but at least its a rack format.
Then we have a few that run virtualisation here and there for some tasks using kvm (some of those too have e350's in them as the e350's do have the virt'n extensions unlike the intel atoms)... we also have a few that run intel based i3/i5/i7 mini-itx boards... they're quite nice when you need some extra grunt...
some others are based on super micro boards as well though (which are quite cheap and run core i3/i5/i7 cpus rather the xeons). Then some others are old 1ru xeons we no longer need for server tasks...
Oh, two other things (and i would test it if i could, but my better half has stolen it). It doesnt require special paper and im pretty sure when the tracking device is plugged in it comes up as a mouse and operates as such (in relative mode)... but dont quote me on that.
I have a CHEAP digital pen (cost me 25$) called the "greenpoint mobile notetaker" (which i think is a pegasus notetaker rebadged).... its ultra simple. it works under linux and what the linux software gives you is a simple svg map of what you drew on a piece of paper. Its just a normal pen with a little tracking unit that somehow tracks everything you write... I dont use it too much, but the times i have its not failed me so far.
http://scratchpost.dreamhosters.com/software/Pegasus_Notetaker/ will pull svg's from the pens tracking device thing
but it looks very much like this http://www.gadgetvictims.com/2009/12/digital-note-taker-pen.html
I find it works ok, but i've not really used any other digital pens, so i have no point of comparison - but at 25$ (which was on sale at the time) i just went "sure why not" and later found out it supported linux (which was a nice surprise).
I had previously looked at things like livescribe and went "no linux support, wont bother". There are one or two i can see on aliexpress http://www.aliexpress.com/item/Free-Shipping-USB-Digital-Pen-Digital-Mobile-Note-Taker-Digital-Handwriting-Capture-Device/519494331.html but i dont know if they're based on the same thing (and they're twice the price i paid) and hence will still support linux
i've seen many ssd's die in enterprise SAN kit and will even go so far as to say as a percentage they die far more often then spinning rusty metal.
However, this is because of how they are used in SAN. Often they are used in a multi-tier way, where most frequently accessed data is pushed up to the SSD's to allow quick access, so they get hit the hardest.
I would be guessing your asking this question simply cause its easier to understand why a part-in-motion can slowly die over time where something thats in silicon shouldnt. You'd probably be surprised to know that alot of drives die a controller death, not a platter/motor death. Typically a platter/motor death usually means a badly made drive (mostly because makers of spinning rusty metal have gotten very good at the mechanics behind them) and while the mechanics in a drive will slowly wear over time, typically something silicon in the controller goes first. The exception to that is where drivers are physically in motion not of their own (i.e. laptops for example), often then the drive shaft in the drive itself starts to get wear unevenly and that usually gets worse over time (or at least, this is what i've been told).
Some SAN makers will even put hdd's through their paces first to make sure they actually perform ok - for eg, they'll measure vibration, etc (i.e. the mechanical components) to make sure a drive is up to spec before it goes into their kit - they cant really do the same as easily with silicon as theres not much to work off that can be measured - so often in enterprise grade SAN's, silicon dies before mechanical.
http://www.makeuseof.com/tag/solidstate-drives-work-makeuseof-explains/ is a fairly good explanation of why ssd's die (and relavent).
Having said all that, i honestly cant wait for the death of spinning rusty metal for the simple reason that ssds should (and havent yet) taken on forms which would be much more useful - why use a sata 2.5" format when you could have much better geometries for example? Then theres the interfaces we use where were really designed with hard drives in mind... but thats an entirely different issue.
As for "if its all the same, why doesnt it die at the same time?"... because at a fundamental level, it isnt the same. When we make anything no matter what it is it, the materials go thru some form of refining process where impurities are removed. Its impractical and near impossible to get anything 100% pure (not to mention entirely uneconomical - you might pay $1/tonne for 90% pure iron and $100/tonne for 91% pure iron as an example). The nature of where we get those materials often means the impurities vary greatly in composition in small spaces of time, hence why two hdd's sitting next to each other on the assembly line might mean one will die after 2 days and the other will die after 200 years. Theres also other components to this in that if you looked at iron (again an example, but true of most metals) under a microscope you'd find that it isnt a uniform substance, its quite grainy so to speak and those grains vary considerably. This too impacts the nature of the substance and its longevity. Using iron as an example, when its cooled its very hard to get a truly consistent cooling of it across the entire piece and the cooling determines how those grains form - i.e. grains in the center of the material will be quite different to grains at the edge and so forth. Thats just a small number of things that explain why consistency isnt quite as 100% as it might appear to be on the surface, there are quite alot of factors that come together to effect materials we use. Ultimately the way we choose the materials we produce things with is by tolerance, i.e. i expect 99% of my metal you sell me to fall into 90% pure and have x tensile strength or any number of variables you might consider important to your manufacturing process but even then its never 100%, you always except at some point that you'll get raw materials that'll fall outside those tolerances and as with everything on the planet its a trade off between price vs quality!
At first i was curious why, but when it comes to the insurance side of things its makes alot more sense...
However, how long until this starts impacting everything you do... obviously in this case things that kill you out-right arent a problem for insurance companies (i would assume). But consider the possible outcomes of such a thing on the following activities:
- scuba diving
- riding a motorbike
- sky diving
- driving a car
- using a 3d printer (printing in abs for example produces bad fumes, depending on what you want to believe)
They are just random examples but all of these have a potential to mess you up in some nasty ways, and in some cases your more likely to end up damaged by them then by smoking (i know a few people who sky dive and not one of them hasnt had some injury thats put them out of action for a reasonable period of time). So before you cheer what appears to be a win for the non-smoking consortium, consider the potential damage to your own after-hours hobbies. i.e. anything an insurance company can say is a "risk" is a potential "sorry, we cant employ you" and that is a rather worrying outcome.
the general role of any company these days seems to be to be able to hide from any form of moral behaviour... the larger the company the more likely they are to behave "within the law" but despicably so. Essentially running a company means you can abandon any form of ethical or moral behaviour - and apparently that's fine.
Patents are a good example of that. They should foster innovation, but they dont for so many reasons. Take apple for example "slide to unlock", "searching the phone AND the internet" these are not innovation, they are simple ideas anyone could have come up with from 5 minutes in front of a phone and yet apple patent them because they can and it gives them leverage to abuse everyone else. In just about any other part of society that kind of behaviour would be considered highly corrupt, yet when it comes to company ethics well there are none and so things like this happen. The criteria for "invention" have become so unbelievable low that simply 3 lines of code might be considered invention if you spin it the right way - for a company this is acceptable, but not from an ethical perspective.
The reality is the patent system should award innovation, but all it achieves is a platform for abuse. From that point alone it probably deserves to die. The problem is that it really wont change much, will apple (and others) suffer terrible losses (which imho, they deserve) from the loss of the patent system? no, probably would work in their favour. Sure they do have patents, but without patents they wouldnt have to pay smaller people for the inventions they license (assuming they do). It might mean people like samsung could take what they want from anyone, copy it, the mass-market it more effectively and just destroy people. That too is wrong.
Sadly, if I working alone came up with some awesome bit whatever that really was "innovation" and "invention" i'd probably end up abused for it. Some large company would simply realise that as a singular individual i probably wouldnt be able to defend myself from their battery of lawyers and so i'd get destroyed anyway. So with or without the patent system, your kinda screwed anyway.
But then having said all that, isnt this exactly what the american capitalist ideal is all about anyway? So perhaps from an american perspective, this is exactly what the patent system was setup to achieve and its doing a great job. I personally just cant see how people can look at the things apple, microsoft, and many others have patented (and used against people) and go "yeah, thats innovation and invention". Its kind of pathetic really.