Archive for the ‘Technology’ Category

Amazon’s Kindle 3: interesting..

Thursday, July 29th, 2010

I’ve had my Sony Reader Touch for close to a year now.  It’s my first electronic reader, and it’s a wonderful device.  The touch screen does add a bit of glare to the screen, though it’s incredibly useful and makes the interface work quite well.

Today, however, Amazon revealed their third version of the 6″ Kindle, with a Wi-Fi only version for $140.  Notably, too, the battery is larger, reportedly lasting up to a month; and they also are packing in a new e-ink display which is supposed to garner better contrast and faster page turns.

DRM improved?

Perhaps I’m a bit late to the party, though I did find on Amazon’s website a certain section labeled “Downloading to Multiple Devices,” which reads in part:

Content purchased from the Kindle Store can be downloaded to your Kindle, or Kindle compatible device, as long as you’ve registered the device to the Amazon.com account that purchased the Kindle content. There is no limit on the number of times a title can be downloaded to a registered device, but there may be limits on the number of devices (usually 6) that can simultaneously use a single book.

That means you can download and read your books on any Kindle device you own as long you’ve registered each device to the Amazon.com account where your Kindle Library is stored.

Personally, this sounds like a big upgrade.  A year or so ago, it was the case that certain titles would only be allowed to be downloaded a certain number of times.  That meant that with multiple devices, generations, etc., the content which you paid for would no longer be yours. Tying the purchases to the account, and having the with “no limit” to the number of downloads sounds like a great step up.

Closing remarks

I personally have no reason to want to stray from my Reader.   It serves me well, is still great for my eyes, and still does have a touch screen which the Kindle series lacks.  Even so, it seems as though Amazon is slowly getting their act together, and I can see that the case may be that they win this e-book stuff in the end, especially if they keep good hardware at such low prices; as well as a reasonable system of DRM to allow your content to actually be your own.

I’m a Systems Ninja

Thursday, July 29th, 2010

Good lord.  Name the number of things that could have gone wrong today, then add five or so for the sake of: reality.

Dell prefab machine went down, so it needs a replacement.  The only sane replacement system has a BIOS password which I don’t know, and a root password I also don’t know.  Only solution: use a root SSH key to access the system on a particular network (static hostname) in order to make required changes.  Boot bare, non-complete system with old, functioning IDE disk, since all other systems are SATA only, in order to find configuration details, including YP, LDAP, NFS, etc. settings.  Make appropriate changes on new system, and move back to a new static IP.  Find extra USB keyboard, since new machine has no PS/2 ports and the only PS/2 to USB adapter on hand wasn’t functional.  Take deep breath; because there was actually a lot more to the situation than the above, though it wouldn’t be good to put those parts on the Internets.

I’m impressed with myself that I got this all sorted out in less than 3 hours.. but good lord did it take some patience and innovative thinking.

AT&T Dataplans, June 2010+: for good ?

Wednesday, June 2nd, 2010

Today, AT&T made an announcement re: some rather significant restructuring of their available wireless data plans.  Namely, rather than an all-in “unlimited” (AKA, 5GB limit) plan for $30/month, there are now two options: 200MB for $15/month, and 2GB for $25/month.

This particular move is potentially a sad change, for those who were previously consuming more than 2GB of data per month on their data plan.  Indeed, for 3GB/month, it’s now $35/month; for 4GB, $45/month, etc.

Realistically, though, I think these changes are for the greater good.  $15/month, from $30/month, means $180/year in savings, assuming it’s possible to stick under the 200MB limit.  On a two year contract, that $180 instead becomes $360.  It’s easy to miss that number when the phone itself is somewhere under $200, a bit more than half of that two year difference.

Changing user habits

Fundamentally, this move is less about the savings which AT&T is trying to ply to its users, and much more about shifting user behavior for their networks.  I have an AT&T account in Boston, and the data network is rather slow; the voice is often unreliable and I get a variety of dropped calls.  Boston, however, seems to be faring much better than San Francisco and New York, where the density of mobile data users is much higher.

Hopefully, what this will signal is a shift towards more and more people using WiFi over their 3G connections.  Indeed, it was reported months ago at Ars Technica that it’s often the power-saving mechanisms of smart phone devices that are clogging the pipes.  Perhaps having the 3G data disabled more often, or relying on WiFi as a primary access method can alleviate these problems.  I’m sure that’s what AT&T is hoping, at least, and it’ll be interesting to see if any of that actually comes to fruition.

Small plug for Symbian

As anyone might know from my previous writings or other general ravings, I’m a big fan of Symbian.  I’m quite impressed by my Nokia 5800′s support for Destinations, which allows for rather granular access rules for managing WiFi and packet data connections.

I went for months with all wireless data disabled from my phone; it took a few clicks, and then there was no way for any application to access my AT&T wireless SIM for any kind of data.  Anything that needed to get online could connect via WiFi, and my wireless bill for data was $0/month.  I then went back to paying for my packet data connection, and still have the option to decide which connection I want to use on a per-application basis.

It’s a highly-configurable system; and aside from the “how do I use this?” factor for certain users, I wish it were more common among platforms.

Final note re: video

I’m not sure why video via packet data connections is as popular as it is.  My guess is that it’s a novel thing for users, and that the YouTube client on the iPhone has been around for years now.  Until LTE or other future mobile technologies come online, I don’t particularly see the point, and there’s a certain Pandora’s Box aspect to it that I wish it hadn’t came up.  In other words, what I’m saying is that large data usage isn’t all that cool, and systems such as the Twitter API which allows for data compression, or systems such as Opera Mini which reduce data throughput are pretty great technologies.  In a sense, the 200MB limit is a nod towards this direction, and it’ll be interesting to see how the market transforms as a result.

Facebook: look, but don’t touch

Thursday, May 6th, 2010

Facebook is absolutely massive.  If you don’t have an account with them, I’d currently recommend that you stay away.  If you have an account, my advice at this point is that while it may not be worth removing your account, it is certainly worth limiting your exposure to the service.

Yes, I said exposure, as though it were some sort of bacterial infection.  There’s a bunch of information on the Internet; the one that caught my eye in a big way was Dan Yoder’s Top Ten Reasons You Should Quit Facebook.  The EFF also has a number of great articles related to Facebook on its Deeplink Blogs related to Social Networks.  I highly encourage everyone visiting this page to dive deeper into these articles, and for those that come in the future.

Look, but don’t touch

My advice to those who aren’t already in Facebook is that it’s best to stay away.  You’ve done alright by this point without it, and there are enough snags that it’s worth keeping away.

For those already in the system, I’m not sure that deleting your account makes all that much sense.  In many ways there is a lot of concrete value in having a Facebook account, connected to your friends and family.

With that said, I do think it’s best to not put any more information into the system, and to remove what Likes you have left.  Think of Facebook more as a read-only system, not as read-write.

Alternatives

In truth, there are countless reasonable alternatives to each of Facebook’s features, and you can rather easily find the same value outside. This may be the time to start a blog, a website, a Twitter feed, a flickr or Picasa account; or, sans all that, to just go back to email as a primary mode of communication.

There are countless other options where you can get out what you put in, without the same threats of privacy or lock-in.  Don’t click on any of the Facebook ads, and if there are people whose content you really like, softly encourage them to put their efforts somewhere else.

I’m not particularly pushing Google, though their Data Liberation Front is a wonderful initiative which puts the power back towards the user.  WordPress, additionally, makes it easy to export all your content in a couple of clicks.  Twitter has an option of protecting your content from anyone beyond the list of those who you accept.

Final thoughts

In the end, I think the solution is by a number of small movements.  With my Facebook account, I may still “Like” someone’s status from time to time, though I’m done with Community Pages or Pages.  I’m no longer letting Facebook import my data from Netflix or Twitter.  I’m no longer going to use its messaging, instant messaging, status updates, notes, wall posts, etc.  It’s up to you to decide where you want to go with all of these decisions, though I certainly do recommend less over time rather than more, in terms of any kinds of Facebook interactions.

re: NYT “Antisocial Networking?”

Saturday, May 1st, 2010

Yesterday, the New York Times put up an article titled “Antisocial Networking?” which I found on their Twitter feed.  It’s an interesting article which covers a few common threads in how our social interactions are modified, either for better or for worse, by the Internet.

I personally feel that face time matters a hell of a lot.  It’s interesting, too, that Facebook is named as it is, when the most of a face that’s visible is a scaled down photograph.  What’s important about the face is in how we figure each other out, and how we pick up on facial cues; as the NYT article mentioned.

In the long term, humans are rather adaptable, as are most creatures.  We can deal with a great deal of a changes, both on the personal and on the larger social level.  In the process of that, however, there certainly are going to be adjustment periods in between, where we falter a bit in adapting to the new set of circumstances.  I almost wish that there was more discussion on the transitions, rather than good/bad polarizations; yet perhaps that hasn’t been studied enough thus far, or it doesn’t sell the story quite as well.

Clearly, access is another large part of the equation.  It’s easy to forget how many people don’t have computers of their own, or even phones of their own, in certain parts of the world.  SMS is certainly huge right now, because it allows for greater access.  With data rates and mobile plans structured the way they are, video chat is most likely going to be a long way off.

MHz for my mobile ..

Thursday, April 29th, 2010

Upon posting my initial thoughts on the Nokia N8, I received a comment which noted, “if it is an old ARM11 rather than ARM8, it will be slow and will fail.”  This hit me on a couple levels, and after typing a lengthy comment I thought I’d instead make my response into a post which may catch some more issues and debate.

Clock speed as a poor indicator

As I see it, the trouble with any MHz war is that it often misses the point of the capabilities of the chips themselves.  Sure, it is rather easy to say that a modern 3GHz machine is going to beat out a Pentium 200MHz, and by that measure we can extract some meaning.  An order of magnitude is a pretty sizable difference, and assuming that processors are made in about the same kind of way there’s a low enough margin of error to assume that one will trounce the other.

In general, though, MHz are not the all singing landmarks of how well a processor works.  Years ago, Intel had pushed their clock rates up and up, even though they didn’t do quite as much per cycle.  This sold more chips, because the numbers were higher and marketed in a way that convinced people they were uniformly better.

AMD, on the other hand, eventually went towards marketing their chips not by the MHz/GHz, but by performance ratings.  I’m not saying that worked out for them in the end, though the reality is that per-cycle, AMD chips generally did more than their contemporary Intel counterparts.

If you want to read further on the subject, Wikipedia has a good article on Clock rate which is worth a gander.

Efficiency is what will drive the future

The next step for processors at this point is less about the clock speed, and more about the efficiency in terms of power usage.  For years, desktop computers have been plugged into the wall, and they haven’t varied all that much in how much power they’ve consumed.  Indeed, because there haven’t always been quite as many options, with Intel leading the industry by leaps and bounds, we buy and use what’s available.  We may shut our computers off or put them to sleep at night, though since the power outlets are almost always available and plentiful, power hasn’t been that much of an issue.

Mobile devices, to include not only phones, but now laptops and tablets of all sorts, are a different issue.  Our computers are getting smaller, more portable, and, most importantly, untethered from the wall sockets.  While battery technology has improved in certain degrees over the past ten years, we are still rather handicapped by modest power requirements.  The trend, then, is that our processors must be efficient: fast enough to do what we need to get done, but also conservative enough to not require bulky batteries or constant recharging.

Back to the Nokia N8, the device is set to have an ARM11 at 680MHz, with a graphics co-processor supporting OpenGL-ES 2.0 (note: for further details on the N8, check out the Forum Nokia website).  D.J., the original commenter, most likely had meant the Cortex-A8, not an ARM8 in his original post; and the Cortex-A8 is certainly a newer chip design than the ARM11.  While it’s probably safe to assume that the Cortex-A8 is a faster chip, I haven’t been able to find any numbers on power efficiency.  These are the kinds of numbers that sound important to me.

Integrated GPUs also important

The second side of the progression of processors is with the GPU, or graphics processor.  Back again to the N8, I found it refreshing that a requirement of Symbian^3 is a graphics co-processor in order to handle graphical tasks.  If I had to pick, I’d choose a longer battery life and boring responsiveness over fancy transitions, animations, etc. in the UI: all of that crap doesn’t make me work any better or accomplish any more.  If the flashiness is going to be there, however, I do think it’s important to be smart about which hardware to use.

In the case of a GPU handling graphical niceties, I then don’t particularly see why I need all that much in terms of a CPU.  The OS depends on lean code that is written properly, and at the end of the day those enhancements will have much greater effect on the reactiveness of the device than the difference between 600MHz and 800MHz.  Symbian may be “old,” though said another way it’s been around long enough to go through iterations and optimizations on a much larger scale.  In that sense, I hope that Symbian does stay old, in certain respects, to avoid unnecessary bloat.  Essentially, having great integer performance can be meaningless in the face of high floating point requirements, and I’m rather confident that the ARM11 in the N8 will be more than enough.

One final note re: GPUs in integrated devices, I find it interesting that the desktop systems are also converging somewhat in this space.  I’m incredibly pleased with my new Intel Core i3 chip, which has a capable, integrated GPU in the CPU die.  The result is that I’m getting 1080p video playback and adequate gaming capabilities without the cost of a video card, in a processor that costs less than $120.  It’s a rather incredible feat, which as time goes on will also find itself in the mobile space more and more as a standard.

Conclusion

All in all, what I’m saying here is that it’s important to take a step back from the tricks of marketing, and to think about mobile devices in many more dimensions than the clock speed of its CPU.  Software, graphics co-processors, and power efficiency are all, in my opinion, more significant indicators of how well a device will perform, and thinking in this way can lead to a better picture about what is actually in play for these future devices.

Less power!

Wednesday, April 28th, 2010

I was originally going to call this post “Kill A Whaat?” though after some thought I figured I’d go with something a bit safer; and less obscure.  By less power, I mean that I’m herein searching out how much power each of my common electric devices use.  In order to derive these numbers, I picked up the P3 Kill A Watt, which I just picked up from Newegg.  The device measures the amount of power that it takes to drive whatever it is you plug into it; rather simple, but incredibly useful.

Fiend that I am for data, I’ve started playing with the numbers right off the bat, and I’m interested already in what I’m seeing.

Desktop computer: Intel Core i3-530, H55 chipset

I put together a desktop in the past month or so.  I’ve kept meaning to write something up about it, though I’ll save that for another day.  My setup includes an Intel Core i3-530 processor, an Asus P7H55D-M Evo motherboard, and an Antec EarthWatts EA-380D power supply; among some other components.

The verdict?  It’s rather amazing in terms of power consumption.  The most power I was able to consume was on the order of 80W, at full load with disk access.  General use was around 50W; booting was somewhere around 60W.  The power factor was rather consistently around 0.90, which is quite good.

In sleep mode, the system consumed 1W, which is the same as with the system off with the power supply not switched off in the back.  In this off state, the power factor was down around 0.12, which gives ~8VA.  That’s not all that much power, though it’s enough that I’m going to shut my power strip off when the system is shut down.

Asus VH236H

I picked up this monitor recently, alongside the above system.  I’m rather stunned at how well it performs, for the amount of money I paid.  Sure, it’s not an IPS, 10-bit display; though it also came with no dead pixels, is 16:9, and fits on my desk pretty well.

36W operating seems to be the standard.  With a power factor of ~0.55, this gives somewhere on the order of 65VA.  In sleep mode, it cuts itself off completely: 0W consumed, which is rather impressive.

Conclusion

I’m not giving a complete list here, in part because I’m not all that excited to tell everyone about every piece of electronics that I own.  You also probably don’t give a shit about what my toaster consumes unless you have the same one, which is rather unlikely.  I will say, however, that working out the numbers can be a lot of fun.