"However, by the time HTML5 is released it's a certainty that the new tags will enhance the search engines' understanding of page content."
This is exactly the point -- these HTML5 tags are being used in production today without people realizing that search engines may not being using the new document outline model since it's a moving spec.
The conclusion is to avoid using these new semantics until they are further developed to avoid an unexpected document outline.
I completely hope that it's the case, and as I mention at the end of the article that could be the case.
Since HTML5 is still a moving spec, it's worth considering how you structure your header tags depending on how much belief you have that this is what search engines are actually doing, given how important the document outline is to SEO.
H2 would be best in a traditional HTML4 docuement, but in HTML5 each sectioning element has it's own document outline. The reason for this is to encourage you to use H1's for the most important content within each section.
In this case the most important content of the <article> section is the title, so an H1 would be best used here.
The point of this post is to show that by using this new guideline, it leads to problems if search engines are using the old HTML document outline.
This is true. A possible solution is if the service was run in EC2, you could leverage the speed on the amazon network if the data was already stores in S3.
I was thinking that as I was typing my comment. Another solution, which S3 uses, is to ship hard disks by courrier. I guess the real metric here is cost per GB transfered in a unit time, say $/GB-hr.
In clusters working on large amounts of in memory data, the approach is often to load the data, then move the code (e.g. a java class implementing some data procesing interface) to the data as required, rather than move the data to the code.
There is always the stuff that goes the other way though like how Seti@Home does FFT's which is computationally expensive and benefits from a distributed system but the file size is quiet small.
Yes, BOINC projects are cases where it is not ridiculous to move the data, because it is computation power that is the scare resource and the work units are typically only in the hundreds of kilobytes to single-digit megabytes.
It seems clear that this was a negotiation point for AT&T to create artificial lock-in on the 3G service, in return they were willing to go month-to-month.
Even if you were able to tether the iPad to the iPhone, AT&T will chart a similar monthly fee. It will be interesting to see if Apple is willing to create separate models to work on CDMA, etc.
"set off the price war by cutting the price of its standard beef bowl to 320 yen, or $3.55"
In the US fast food restaurants all include a "99c" menu. You can get burger, fries and soda for $3. Seems like $3.55 isn't a shocking amount for a leading fast food chain.
Certainly the margins on a beef bowl are different than a burger and fries, but it struck me that these amounts aren't as low as I expected give that it's becoming national news in Japan.
I live in Tokyo right now. Tried to setup a HN Tokyo GT, but that didn't work out :/
You can eat at McD's for 100 yen, maybe 350 yen for a happy meal. I used to get by on McDs during my ultra poor initial few months in Tokyo.
As for eating properly; I usually eat for about 1000 yen when I go out, as I know a lot of the decent / cheap places. However, step into any reasonable restaurant and you're likely to be hit with sticker shock. I don't think it is really possible to eat healthy for less than 2000 yen per person without having extensive local knowledge.
As for the deflation story..
Commodities and groceries cost more than they did 2 or 3 years ago. Cheese, for example, costs almost twice what it did when I first arrived in this country (2.5 years back). Living space is at a premium in Tokyo. The apartment renting scene is as shark-infested as ever. The monthly price was more or less the same as it was last few years, and they basically wanted 6 months pre-payment for anything built since 1980.
The Japanese media has been hyping deflation, but the only deflation I see is in salaries. Whether it is anime artists getting by on as little as 100,000 yen a month in Tokyo (!) or my girlfriend working 15 hours a day without overtime (for Proto corporation, the media conglomerate that did Goonet), it seems everyone "normal" is making less. (This might be quite different outside Tokyo, as most other prefectures have been losing population this entire time.)
As for myself? Oddly the economy is picking up for me.
Every time you select something they post back to their server with details about the selection (including the article, location of your selection, referrer, etc)
That's already the case for AJAX requests in most browsers. The problem is that Tynt doesn't use AJAX for the communication (they can't, because they communicate with a domain other than the one serving the content). I'm guessing they use a hidden iframe instead.
This is exactly the point -- these HTML5 tags are being used in production today without people realizing that search engines may not being using the new document outline model since it's a moving spec.
The conclusion is to avoid using these new semantics until they are further developed to avoid an unexpected document outline.