WebLogic Server (WLS) およびJavaでは様々な方法でスレッドダンプを取ることができます。その方法の詳細を以下にまとめました。スレッドダンプ採取にあたっては、Javaクラスや管理コンソールではなく、OSコマンドを使うことを常に推奨します。それは、管理コンソールがハングアップした場合、ユーザーが接続できずスレッドダンプを採取できなくなるからです。
Java Mission Control、Java Flight RecorderはJava SE Advanced、もしくはJava SE Suiteをお求めの方がご利用いただける機能です（つまりお金必要ですよ、ということですね）。 とはいえ、ちょっと動作を試してみたい、ということもあるかと思います。その場合、上記スライドにもある通り、個人的な評価であればBCLライセンスに基づいて利用可能です。利用に当たっては、次のコマンドを追加する必要があります。
Microsoft started with the flat Metro interface of Windows 8. Apple flattened iOS 7 and now Google has flattened their logo.
Is this based on human-computer interface research or is it style and fad? Years ago, I taught an HCI course in which we read about controlled studies in making user interface design choices. Google is said to be engineering driven, researching everything from the optimal number of seats at cafeteria tables to the number of pixels devoted to a particular icon on the screen.
But, I have the feeling that the rush to simple, flat user interfaces is driven by fad as much as function. My wife uses an iPhone and an iPad. She is probably a typical non-geek user -- using relatively few apps and features. Our daughter was all excited when iOS 7 came out -- she grabbed my wife's iPad and iPhone and installed it the next day.
After using iOS 7 for a few days my wife says she sees no advantage for her usage mix -- the only change is that she has to learn how to do what she always did in slightly different ways. She also likes 3-D buttons that are obviously clickable and give feedback when you click on them.
If these changes were driven by functional considerations, I'd love to see the old-fashioned HCI research that led to them. If not, fads tend to be cyclical, so we can look forward to a return to 3-D skeuomorphism in a few years. As you see below, the designers are already working on them.
(You can make your own goofy logos at http://createfunnylogo.com/).
-----
Update 9/29/2013
This child was even more upset than my wife with iOS 7:
Posted by UnknownJumat, 27 September 20130
komentar
I give my students feedback as to how they are doing relative to the class on assignments and quizzes throughout the semester. I also do informal, anonymous surveys to give them feedback on their relative level of effort, for example asking how many many actually watched a video that had been assigned or how long they spent studying a module or doing an assignment. These low-tech surveys take only a few minutes to administer and tabulate. My goal is to put information, and therefore responsibility for the outcome, in their hands.
Purdue University has a more ambitious effort for monitoring student progress and giving them feedback during the term. Since 2007, they have used Signals, a system that mines student data to predict their success in a course. They look at demographic, help-seeking and performance variables to judge how well a student is doing as the term progresses. Student feedback includes a red/green/yellow signal to indicate their overall progress as well as "canned" emails from their instructors and suggestions as to how to improve and where to get help.
Purdue reports that graduation rate, retention rate and grades have improved for students taking Signals courses. For example, Signals students got better grades in these courses:
This and the other results reported by the Signals team are encouraging, but there are many confounding variables. For example, professors choose whether or not to use Signals in their courses and those who choose to do so may be more more committed to teaching than those who don't.
I am not sure whether I would use Signals if it were available to me as a teacher, but my low-tech approach to feedback on relative performance and effort is simple and I can do it on my own.
-----
Update 9/28/2013
Andrew Stewart suggest that we look at JISC's work in this area, writing:
Posted by UnknownKamis, 26 September 20130
komentar
In an earlier post, I discovered that my old laptop was too slow to cast video to my TV set, but it worked well when displaying still images like a PowerPoint presentation on a TV set equipped with Google Chromecast.
In the 1980s, local area networks became common, and the University of Arizona, Xerox Palo Alto Research center and others began experimenting with augmented meeting rooms with shared displays. Meeting participants could brainstorm ideas, rearrange document outlines, edit documents, vote and conduct polls, etc. and companies like Groupsystems marketed upscale decision support rooms like this one.
Those rooms cost a fortune and were used as corporate boardrooms, but interest in them has waned.
A large TV set with a Chromecast dongle is not as powerful or opulent as a decision support room, but for $35 it might become the decision support room for (small groups of) the rest of us. Below you see a simple test in which two users are editing the same Google Drive document with the result displayed on a TV set.
This sort of setup would allow several people in a room to share a heads-up display. I tested it with a simple Google doc, but one can imagine using it with the kinds of software found in expensive decision support rooms -- software with modules for voting, brainstorming, outlining, writing, etc. and features like selective anonymity and podium passing. We may find Chromecast displays next to those whiteboards in our offices in the future.
----- Update 10/2/2013
Commenter Martyn Williams noted that the Chromecast will work with an HDMI-equipped projector, which would enable larger displays in your office or conference room.
Commenter Roger Jennings suggested that the Chromecast could save bandwith in casting pages if it had more "intelligence" -- for example being able to display PDF, Excel, Powerpoint, and Word files.
Together, these comments suggest a future where we have dumb displays with upgradeable, external intelligence. I have changed the TV set in my den 3 times in the last 35 years, but change computers every couple of years. I don't want a smart TV set, I want a dumb display with an upgradeable "chromecast."
Futurelearn, a coalition of 20 UK and international universities, the British Museum, British Library and British Council, has just launched with 20 courses starting this fall and winter. The courses are 6 or 7 weeks long and require 2 or 3 hours per week.
I checked out their "beta" Web site and few things caught my attention.
They also show interest in training for job-related skills. One of the initial courses is Dental photography in practice.
Their tagline is Enjoy free online courses from leading UK and international universities, indicating a global focus. In addition to international universities, they will be serving international students. One of their partners is The British Council, the UK international cultural organization, which offers classes (online and off) and arranges cultural and educational exchanges and events. The British Council has offices in 116 nations and they will no doubt help with marketing and spreading the word.
Futurelearn is later to the game than the big three U. S. MOOC providers, but the game is just beginning -- the technology, pedagogy and place in society of online education are all changing rapidly. Furthermore, FutureLearn is a private company wholly owned by the Open University, which has been doing distance education (online and off) since 1971.
I've enrolled in a course and am anxious to see their platform and pedagogy. Stay tuned.
I heard a rant by Jeff Jarvis on the This Week in Google podcast. It seems that he got a new Nexus 7 tablet and Verizon refused to add it to his LTE account because it had not yet been verified. He tested it with a SIM from a different device and it worked fine. He also pointed out that Google had advertised that it would work on the Verizon network and that the terms of Verizon's FCC license required open access to any compliant device.
Verizon said they had to certify the device -- have it tested to be sure it would not harm their network.
That reminded me of the Hush-a-Phone. In 1956, the courts overruled an FCC ban on Hush-a-Phone, rejecting AT&T's claim that it posed a risk to the network and would degrade call quality.
Here is a picture of the Hush-a-Phone -- you can decide how grave the risk was:
What if AT&T had prevailed in the Hush-a-Phone case and the subsequent case of the Carterphone, a device for patching radio calls into the telephone network? (Yeah, hams used to do that).
It seems that Verizon is unclear on the meaning of "open" -- they are still nostalgic about the good old days, when only the phone company could sell you things like phones, modems, DSL routers, answering machines, etc.
Posted by UnknownJumat, 20 September 20130
komentar
A lot of folks are saying that phones are so fast these days that the 64-bit processor in the iPhone 5s does not change the user experience -- it is nothing more than a marketing gimmick.
Wrong.
This reminds me of the time I was consulting to MicroPro International, publisher of WordStar -- the most popular word processing program of its time. I was fired as a consultant after telling them that the second-generation word processor they were developing, WordStar 2000, was a loser because it was not graphically oriented. They ignored me because, hey, a graphically oriented word processor would require 64 MB of memory and 16 MB was a lot of RAM at the time.
In a Moore's Law world, you design for the future, not the present. Apple's 64-bit CPU will not make placing phone calls or texting any faster, but it provides a new platform on which to build new applications. For example, we will see the substitution of computation for hardware in making better videos and photographs -- the 41 megapixel camera in the Lumia 1020 will not seem so amazing in a few years. Improved voice recognition and synthesis will demand more horsepower. Wearable things and physiological monitoring applications will be developed to use all the processing power we can muster.
And, how about something mundane like PC replacement?
I do 99% of my work on that machine, and unless I am streaming or rendering video, it seems quite responsive. If the CPU were ten times as fast, I don't think I would notice much increase in my productivity.
The iPhone is faster than other phones, but how would it compare to my laptop if it were plugged into a keyboard/monitor docking station?
Both have dual core, 64 bit CPUs and judging by the transistor counts and cell phone benchmarks, the iPhone processor should be able to beat my laptop. But, what about the low clock speed and relatively small memory? And two cores sounds kind of lame these days.
Apple has kept clock speed, memory capacity and core count low to save power, but, when docked, power is a minor consideration. I bet Apple's clever engineers could design a dual mode machine, that slowed and saved power when not docked. (Cooling would be a problem when docked and running fast).
Canonical is taking a shot at cell phone docking with Ubunto for Android. It is vaporware for now, but they have a cute video to illustrate the concept:
Apple traditionally does a major iPhone upgrade every two years. This was the year for a minor upgrade, but it laid the foundation for the future. I do not know what applications will be developed to utilize that 64 bit address space and processing power, but ... build it and they will come. ----- Update 9/21/2013 Last week, I asked my class whether they thought the finger print reader on the new iPhone 5s was a big deal. They did not think so -- they said they would be willing to pay from 0 to about $10 for the added convenience. Then I asked them about using it as general ID for authentication and for purchases in stores and online. That perked their interest up. But, will that happen? Brian Roemmele thinks it will and builds a strong case in his post What is Apple’s new Secure Enclave and why is it important?. It turns out that some of those billion plus transistors in the the A7 CPU are devoted to the implementation of patent protected security features.
If Roemmele is right, we will see a slew of authentication and transaction-oriented applications for the iPhone 5s and future devices using this technology. Here is the conclusion of his post:
Apple has taken a very slow and methodical approach with the release of Touch ID. We can see that there was a tremendous amount of amazing work that has gone into this project. All of this convergence took over seven years of very hard work. It includes many patent applications, the acquisition of AuthenTec, the selection of the A7 processor and the integration of the TrustZone suite all baked together into what we now know as Touch ID. This has been a long journey that has only just been made public and I am rather certain that Steve Jobs would be quite proud.
-----
Update 10/22/2013
The new iPad and iPad mini are out and Gigaom's first look singles out the speed of 64-bit apps and predicts that other developers will follow suit. Here is a quote:
The native Apple apps open super quick and there’s no lag when scrolling or paging through content in Pages or iPhoto. You tap and the device responds. Obviously Moore’s Law is at work here, but it helps that Apple has rewritten its native apps for 64-bit compatibility to be fully optimized with the A7 chip. Developers will be doing the same over the next year, so the iPad Air is likely only going to get better until the next model arrives.
The survey showed that 38% of the people in the U. S. subscribe to or watch Netflix streaming video service. That is up from 31% last year.
I don't know about you, but that is a lot more than I would have guessed. It is about 119 million people if they consider the entire population -- babies and all. Note also that Hulu and Amazon also have significant, growing numbers of subscribers.
Netflix' view of themselves as a series network is also confirmed by the study. Forty five percent of Netflix streaming subscribers say the types of shows they watch when they stream are original programming -- series like "House of Cards."
And, when they watch those series, they tend to “binge.” Eighty eight percent of Netflix users and 70 percent of Hulu Plus users report streaming three or more episodes of the same TV show in one day. As we pointed out in our previous post, both consumers and creators like the full-season format of Netflix productions.
The survey also showed, that Netflix and Hulu are watched on a variety of devices:
The above figure also suggests a trend away from computers and game machines toward phones and tablets. People want to watch TV on any device at any time and at any place.
"Over the top" Internet television is not just for geeks any more -- 38 million people understand how easy it is to defect from cable and satellite TV, to "cut the cord." As the quality and variety of Internet TV material improves, it will be easy for them to drop their cable and satellite subscriptions. When we reach the tipping point, the transition will be rapid.
Posted by UnknownRabu, 18 September 20130
komentar
Check out the Netflix Long Term View document for investors. It is interesting because it reveals their vision of the company and for the future of television in general.
But, they are not just a network, they are a "TV series network." That is a testament to their success producing multi-episode series for Internet distribution. People like to watch TV without commercials. They also like watching two, three or maybe all the episodes in one sitting. They like watching TV on phones, tablets, PCs or television sets whenever they want to.
Creators also like the artistic freedom and financial security of the Netflix format. They have the freedom to craft a long story. We can think of the Netflix series "House of Cards" as a 661 minute story, to be watched in one or several sittings. The 13 episodes vary from 46 to 56 minutes in length -- the writers are not constrained by time slots and commercial breaks. They could have created more or fewer episodes if they that was the best way to tell the story.
For more on a creator's view of Netflix, watch Kevin Spacey's excellent 2013 James MacTaggart Memorial Lecture 2013 (below). Spacey, who produced and starred in "House of Cards," says he pitched it to the TV networks and cable companies, but he went with Netflix because they gave him total creative control, funded 26 episodes up front and did not request a pilot episode. He feels that pilots are an expensive, misleading digression. He is producing very long stories with evolving characters, and trying to cram that into a single TV pilot or a two hour movie is impossible.
How about the Netflix vision of the future of television?
They sum it up saying that "While Internet TV is only a small percent of video viewing today, it will keep growing because:"
The Internet is getting faster, more reliable and more available;
Smart TV sales are increasing and eventually every TV will have Wifi and apps;
Smart TV adapters are getting better and cheaper;
Tablet and smartphone viewing is increasing;
Internet TV apps are improving through competition and frequent updates;
Streaming will be the leading source for 4k/UHD video;
Internet video advertising is becoming more personalized and relevant;
TV Everywhere provides a smooth economic transition for existing networks;
New entrants like Netflix are innovating rapidly and driving improvements.
Do you like "binging" on a commercial free TV series? (I just finished "Orange is the New Black"). Do you agree with Netflix' reasons for the future growth of IPTV? Are they enough to overcome the power of the incumbents?
This post is a bit unusual for me, but it fits because the Internet is the product of business and entrepreneurship as much or more than it is of research and education.
O'Reilly Media is one of earliest Internet companies. O'Reilly started as a technical writer then began publishing books on the Unix operating system. By the time the Web appeared, the company was publishing Internet and programming books and they built the first commercial Web site, which was later sold to AOL. Today O'Reilly Media is a profitable group of enterprises with nearly 500 employees and collective revenues approaching $200 million.
Founder Tim O'Reilly has written a longish blog post in which he recounts the history of the company and talks about what he did right and what he did wrong. He modestly features six key mistakes and the lessons he learned from them, but there is a lot on what he did right as well.
JDKにバンドルされる商用機能のJava Mission Control (JMC)は、Javaアプリケーションの監視・管理・プロファイリングのためのツール群です。ガイドドキュメントにはJava Mission Control User's Guide.は含まれています。Java Mission Control Release Notes.もご覧頂けます。
「Language Tag Filtering and Lookup」という新しいページがJava TutorialのInternationalizationの項、Setting the Localeに追加されました。このページでは、JDK7で導入された言語タグ、言語範囲、言語優先度リストを説明しています。これには言語タグのフィルタリングと言語タグのルックアップに関するセクションも含まれており、コンパイルして実行できる多くのサンプルがあります。
この機能強化にはSHA-224、SHA-256、SHA-384、SHA-512メッセージダイジェストと、SHA224withDSAとSHA256withDSA署名の追加が含まれています。また、Also, the keysize limit for the Diffie-Hellmanアルゴリズムの鍵サイズの上限が1024bitから2048bitに増加しました。
Java SE 8で、Javaクラスファイルフォーマットがアップデートされ、JSR 292（Supporting Dynamically Typed Languages on the Java Platform）によってもたらされた変更を反映しています。これにより、pack200エンジンはクラスファイルを効果的に圧縮するように適宜更新されてきました。特に、pack200エンジンは定数プールエントリおよびJSR292で導入された新しいバイトコードを認識するようになりました。その結果、pack200ツールのこのバージョンで作成された圧縮ファイルには旧バージョンのunpack200ツールとの互換性はありません。
