目前会议日程已经公布在年会自己的注册系统上。在这里注册的参会者，可以点击进入自己感兴趣的话题，标记上你是否对此话题感兴趣。这样在场外，如果还想继续讨论自己感兴趣的话题，可以方便寻找其他对此话题感兴趣的blogge，以便进一步探讨和交流。

下面是目前拟定的会议日程的拷贝，仅供参考。这个日程仍然可能会有所变动和修改，请以会场所发的会议手册为准。

活动议程

2007-11-03, Sat

• 开幕 主持人: 陈涓玲 潘剑峰 ( 09:00AM - 10:00AM )

• 微Blogging和潜力 主持：潘力 ( Li Pan ) 嘉宾：Alex Mou* ，Livid* ，Jiwai.de* ，林凱洛 Carol Lin ( 学学)* , 饭否 团队 ( 10:00AM - 11:00AM )

• 社会性软件与第一人生 主持：Luyi Chen 嘉宾：陈浩源* ，郑晓云，Alen Tien*，郑高照 ( 10:45AM - 11:30AM )

• 视频社会化服务 嘉宾：Wang Wei ，56.com ，6rooms ( 11:30AM - 12:30PM )

• 播客在中国 演讲人：林嘉澍 ( 01:30PM - 02:15PM )

• 研讨：草根媒体与专业媒体 主持：Rebecca Mackinnon 嘉宾：闾丘露薇* ，周曙光* ，翟明磊* ( 02:15PM - 03:30PM )

• Map服务 嘉宾：庞小伟(e都市)* ， 叶群峰 (网络天书)* ，Mapbar ，余建军( 城市吧 )* ( 03:30PM - 04:15PM )

• 社会协作力量如何改变翻译格局 演讲人：张雷 ( 04:15PM - 04:45PM )

• Wiki的创新应用 演讲人：潘海东* ( 04:15PM - 04:45PM )

• Wikipedia 在中国 主持：时昭 * 嘉宾：Andrew Lih，苑明理* ，许瑜真*，潘海东*，等 ( 05:15PM - 06:00PM )

2007-11-04, Sun

• Keynote: Sharing Economy 演讲人：伊藤穰一 Joichi Ito, Chairman, Creative Commons ( 09:00AM - 10:00AM )

• Music 2.0:社会化音乐分享的矛盾与价值 主持： 汤永钢* 嘉宾：杨勃，Allen Guo* ，David Ding* ， Zhang Youwei 等 ( 10:00AM - 11:00AM )

• 口碑营销和商业底线 主持：Kaiser Kuo 嘉宾：Keso* ，Jim Sang* ，吕欣欣* ，刘新华 ( 11:00AM - 11:45AM )

• 玉米 2.0 — the latest trend in domain name industry & traffic monetization 演讲人：乔敬(Ching Chiao) ( 11:45AM - 12:30PM )

• 艺术1.5:”电子”艺术 主持：Ajiao * 嘉宾：Sun Tian ， jeuce (skype from Australia)* ，李如一 * ，黄奎 * ，王晓峰 ，胡戈等 ( 01:30PM - 02:15PM )

• 互联网时代的NGO 演讲人：张利 (vim) ( 02:15PM - 02:30PM )

• 互联网时代的NGO——了解和实现IT/WEB2.0对于社会性工作的影响和价值 主持：张利 (vim) 嘉宾：安猪 ，胡昕 ，慈善一加一 ，李明顺 ，严天亦 ，Joy Tang* ( 02:30PM - 03:15PM )

• 社会性学习 演讲人：庄秀丽 ( 03:15PM - 03:30PM )

• 教育与Blog 主持：余扬 * 嘉宾：吴向东* ，KenLee* ，framces*，Qienkuen*， ( 03:30PM - 04:15PM )

• Open Source Business 主持：王鹏嘉宾：Robert Scales* ， Kris Krug * ，黄东* ，徐继哲* ( 04:15PM - 04:45PM )

• 西方学者眼中的中国博客圈及他们为什么错了 演讲人：吴迪 ( 04:45PM - 05:00PM )

• 第二人生与商业模式 主持：蒋昌建 嘉宾：虚游记，安舍钟 ，等 ( 05:00PM - 05:10PM )

• 个人门户的应用价值 演讲人：卢刚 ( 05:10PM - 05:20PM )

• 电信服务与网络法律 主持：Mountain* 嘉宾：Yetaai ，刘晓原 等 ( 05:20PM - 06:00PM )

Microslow patch is used by many DBAs and developers to accurately time their queries and to catch those which run less than a second as they can also be a performance killer for a busy application.

Recently I have started the development of an updated version of the patch. The basic idea is the same as for its predecessor - to get more information about query execution logged into slow log, however the new version is loaded with a set of cool new features.

CONNECTION IDENTIFIER

Each slow log entry now contains a connection identifier, so you can trace all the queries coming from a single connection.

MICROTIME RESOLUTION QUERY TIMING

This is the original functionality offered by Microslow patch. The new edition is free of a tiny bug which was to treat long_query_time value as seconds. In effect one could only choose to log all queries to see those which take less than 1s. Now long_query_time is in microseconds and you can set it for example to 300000 which is 0.3s!

LOGGING OF THE REPLICATED STATEMENTS

Normally MySQL will not write into slow log any queries executed by the slave’s SQL thread. This patch allows you to change that behavior with the new parameter --log-slow-slave-statements.

THE DETAILED INFORMATION ABOUT QUERY EXECUTION

Each query can be executed in various ways. For example it may use indexes or do a full table scan, or a temporary table may be needed. These are the things that you can usually see by running EXPLAIN on the query. The patch will now allow you to see the most important facts about the execution in the log file.

QC_Hit tells whether query cache hit occurred or not. If it says ‘Yes’, then all other values will be ‘No’ or 0, because the query was not actually executed by the engine.

If Full_scan is set then most likely the query is a bad one, because it read all the rows from a table.

Full_join means any of the joins didn’t use indexes.

If a temporary table is required for query execution Tmp_table will be set. Sometimes that table must be created on disk instead of in memory, in such case Disk_tmp_table will state that.

Filesort is for indicating that filesort algorithm was used, while Disk_filesort means that sorting was done through temporary files.

INNODB USAGE

The final part are the InnoDB usage statistics. MySQL currently allows you to see many per-session statistics for operations with SHOW SESSION STATUS, but that does not include those of InnoDB, which are always global and shared by all the threads. The patch gives you the opportunity to see those values for a given query.

InnoDB_IO_r_ops is the counter for the number of page read operations scheduled. The acutal number of read operations may be different, but since this can be done asynchronously, unfortunatelly there is no good way to measure that. InnoDB_IO_r_bytes is the corresponding value in bytes.

With InnoDB_IO_r_wait you can see how much time (in seconds) it took InnoDB to actually read the data from storage.

InnoDB_rec_lock_wait is the time (in seconds) query had to wait on row locks.

InnoDB_queue_wait is the time (in seconds) query spent either waiting to enter the InnoDB queue or inside that queue waiting for execution.

InnoDB_pages_distinct tells the number of unique pages accessed by the query. This is actually an approximation based on a small hash array representing the entire buffer pool, because it could take a lot of memory to map all the pages. The inaccuracy grows with the number of pages accessed by a query as there is more chance for hash collisions.

In case InnoDB is not used for the query execution that information is written into the log instead the above statistics.

The full slow log entry for a query using InnoDB can look like this:

or if InnoDB was not used:

Important note: Although the patch was tested and run successfully under stress with both synthetic benchmarks and real database traffic, it's considered experimental and as such I don't recommend using it in production or on any important system!

The patch is available for MySQL Community 5.0.45.

Maciek Dobrzanski

Entry posted by Maciej Dobrzanski | No comment

Add to: | | | |

6A: This really depends on your community and the conversation at hand. Inside an organization, if you're blogging on an intranet, people will generally be pretty civil because of the social pressure to act appropriately around coworkers. On a public blog, you can often keep even a contentious conversation from becoming uncivil by simply requiring some form of identification from your commenters. But platforms like Movable Type have very robust tools for moderating conversations -- if you see a topic that's really becoming heated, you can always set comments to be moderated by default, and then publish only the comments that are appropriate. It's a lot easier to err on the side of being more restrained and then open up over time than it is to try to shut things down after the fact. For more, see "Control Your Comments: It's Your Lobby", or consult the Six Apart Guide to Business Blogging, which has an entire chapter on creating a commenting policy.

6A: Blogs and wiki software are very complementary tools, with slightly different goals. Platforms like Movable Type give you the rich ability to easily update web pages, host conversations, store media assets, and create standalone web pages, all within the easy experience of a blogging platform. Wikis are a great supplement to these kinds of tools if you want to add in the ability to have a number of people collaborate on a document to reach a consensus. In contrast, blogs are a great tool for providing status updates or for discussing topics where multiple individual viewpoints can be useful.

One major social consideration in deploying these tools is that since most wiki tools allow anyone to edit a document, contributors can be concerned about getting appropriate credit (or blame!) for their contributions, and discussions where people have opposing viewpoints can become difficult, resulting in successive edits to replace other people's content. Those caveats aside, wikis and blogs enhance each other's functionality, and page 9 of the Business Blogging Guide offers a more in-depth compare-and-contrast look at the two technologies. If you're looking for a platform that integrates these tools, you should consider SuiteTwo, an Intel-led Web 2.0 suite for the Enterprise which incorporates wikis as well as using Movable Type as its flagship best-of-breed blogging component.

6A: The right techniques for promoting a blog vary widely, depending on what audience you're trying to reach. Inside an organization, you should take advantage of the other channels that people are already using to get information. Include links to your blog in company newsletters, list your blog's web address in your email signature, and display recent headlines from your blog on your company's intranet homepage. (Most portal software applications let you include RSS feeds pretty easily.)

If your'e trying to reach an audience outside your company, you'll want to use all the traditional methods for promoting any communications channel. Make sure all of your staff members who talk to customers or the public know about your blogs and are comfortable describing them to others. Don't be shy about including your blog's address on your business cards, in any materials that are included with your products, or even on direct-mail communications or in print ads. Every channel that you reach your audience with now has the potential to become a two-way ongoing conversation through your blog.

争夺Facebook的竞争中，Google暂时败给了Microsoft, 于是便抛出了OpenSocial, 不能不让人怀疑Google是用OpenSocial来“搅局”的。

Open Social究竟会是什么样子，目前google之外还只有少数人知道，从ZDNet上一篇blog来看,  Open social将不仅仅是一套API和协议，而是一个network computing platform, 这样和facebook app开发相比，开发者投入更小，数据可以完全hosting在google平台上，如果万一造就了一个人气很旺的killer apps, 根本不用担心没钱买硬件和带宽了。

从Techcrunch的报道看，google联合了Friendster, Linkedin, Salesforce.com, Oracle, Hi5, Plaxo等全明星阵容， 如果真的如同所预期的那样能通过OpenSocial API来获取这些平台上的用户数据、social graph(用户之间的关系)、activities (应该理解为比较广义的事件，例如用户发表了文章、post了图片 etc.)的话，那实在是很令人激动的事情。 此外OpenSocial还将采用标准的HTML, Javascript作为开发语言，而不像facebook, salesforce那样定义自己的改进的mark up language, 使得基于此的应用独立性更强。

看到这样的梦幻阵容，我不禁想起了当年Microsoft 发布passport的时候的场景，那时候passport的支持伙伴何尝不是梦之队呢？ 如今passport的支持者仅仅只有微软自己。 并非passport技术有问题，并非用户无此需求，一切都是商业利益权衡斗争的结果。 有人说passport不能广泛支持是因为数据由Microsoft一家独大持有导致的，但走开放、对等的openid的发展不力现状似乎让这些说法并不能站得住脚。

OpenSocial这样的概念其实也并不新鲜和革命，只不过在过去一样只有创新者才会提出这样的思路，而创新者缺乏的不是思路和技术，而是资源。现在google也许能有这样的资源来把OpenSocial真正做起来了。 然而OpenSocial最终的成功与否不能只看google, 还得看他的那些合作伙伴是否能真正地提供配合，以及开发者社区是否能在此之上建立丰富而有价值的应用。

几家欢喜几家愁，这边google, Microsoft, facebook正干得热火朝天，那边Yahoo 360 宣布明年要“重组”, 被很多人解读为“关闭”。yahoo 360是一个中规中矩的应用，功能很强、设计也不错但开放性却相当不好，加之和yahoo收购来的众多服务存在重叠，因此发展不温不火也实属意料之中。我记得去年参加Yahoo hackday, 一个获奖的是个女孩子，她hack的内容就是给yahoo 360换了个类似myspace的skin, 变得花哨不堪，她的发言大意是Yahoo 360是个爷爷奶奶级别的用的sns, 所以她给360带来了新活力，台下掌声雷动结果她获了一个大奖。

现在评论什么还为时过早，过几天google正式发布了之后试用后再作评论。 不过需要知道google不是万能的，google有Google Map API的成功，也有出来时一片叫好但出来后被人遗忘的如Google Set, Google base等同样看上去很美、很先进也很“开放”的东西。

不过可以相信，开放的web应用平台的趋势已经越来越成为主流，

延展阅读：

http://www.codingmill.com/forgotthemilk/articles/third-chat-about-sns-google-social.html

http://www.nytimes.com/2007/10/31/technology/31google.html?_r=1&oref=slogin

Share This

Check out these presentations from Oracle OpenWorld Asia Pacific that was held in Shanghai this year. By the way, I got my Oracle OpenWorld badge a few days ago. That will save me some time during registration. Apart from the sessions, here are some of the things that I’ll be doing during the conference (so far):

• Sunday afternoon, briefing with the database product team followed by the Oracle ACE dinner.
• Monday 9:45 AM, podcast with Justin (unless it’s rescheduled).
• Monday 6:00 PM, AppsLab meetup.
• Monday 7:30 PM, OTN night.
• Tuesday 7:00PM, blogger meetup.
• Wednesday 8:00 PM, the OpenWorld Festival.

It’s going to be loaded with learning, networking and face-to-face socializing. Fun, fun, fun.

---
Related Articles at Eddie Awad's Blog:

• Oracle Open World 2006 - Download Presentations Now
• Oracle Database 11g Overview Presentation
• A Complimentary Oracle OpenWorld Session
• New Oracle PL/SQL Error Management Framework Released
• Oracle OpenWorld News Aggregator

杜琪峰十分多产, 作品的品种也十多元. 喜剧爱情黑道警察鬼片的各种港产片典型都有, 每个时期都有其代表作.

天若有情的港式草根感情美学, 东方三侠的港曰漫画动作悲情大结合, 与周星驰合作的夸张神怪喜剧, 韦家辉刘德华郑秀文联合的大众爱情电影男女系列, 警察与黑道的枪火美学, 暗战暗花的黑色男性美学. 杜琪峰是少数在商业上和创作上都能找到观众和风格的电影导演. 杜琪峰没有王晶那种玩世不恭那种以搵食为荣 , 杜琪峰是一个认真的导演. 在众多八九十年代的男大导之中, 杜琪峰仍然有着一份对电影言志的热情, 在创作上也在努力推陈出新, 寻找一种港产片的大男人美学.

香港的男性导演都流着一种大男人美学和价值的血, 杜琪峰更是一种典型. 和韦家辉合作的时期, 大男人好象被韦家辉柔化了, 剧本的语境比较文艺, 剧情也少是以前的单刀直入, 多了在人物和情节的复杂情. 但主调都是一种男性的暴力美学与及对正义的一些坚持. "真心英雄"是一种极致, 刘青云和黎明在那一连串的暴力设计, 那种比埕伦天奴的暴力美学还暴力的设计, 带出了一个比童话更童话的男性暴力童话, 片没那一种夜总会大血战, 黎明的冷面与青云的冷静, 是港产片极少出现的一种类型片示范表演, 也可以说的社韦组合作品当中最完满的一次合作.

杜琪峰的局限也是港产片的局限, 港产片的那种缺乏自我反思的动物主义, 成为了一种不段被复制的方程式, 杜韦的合作比较有意思就是带有多点反省, 但那都不是自觉的, 大只佬和尚是有着一种突破, 以佛法的观念说明暴力的本质, 但后半部好象太过要把佛理说明, 而失去了让观众体会的可能. 最大的原因也是港产片那产画公仔昼出肠的作谓要观众明白的本质, 没有留下什么可以思孝的空间, 逃不掉港式娱乐片的格局.

枪火是小品, 是一枪风格化的表演, 简约的对白, 简单的剧情, 鲜明的人物设计, 有着一点点黑泽明七侠四义的影子, 在娱乐与风格之间找到了一种模式, 像现代舞一样的动作设计, 带来一种不是暴力的暴力美感. 也是一种曰式动作漫画的质感. 后来的放逐失去了枪火的神采, 人物多了, 故事复杂了, 人物与人物之间的关系是太多的设计, 枪火的动作没有一种生命力, 结局的大枪战好象是有点失控. 加入的女性角色没有什么明确的心理论述, 像陈雅伦的妓女角色只沦为一种情节的借口, 何超仪的妻子角色的简单也是难以致信.

韦社各自创作之后, 杜琪峰进一步开展其黑色电影美学, PTU 也是以"枪火"的影像主导作为叙事主体, 以小人物的角度带出一种黑色黑道的道德观与及事件发生的不可预知性. 林雪的小人物与任达华的为了原则的一错再错, 与结局的各种不预知的事件也是非常特然的一种设计, 大团圆式的结局只可以说是林雪小人物的傻人有傻福的结果, 也是一种观众期望看见的结局类型. 也是社琪峰影片结局的一种典型.

小津安二郎一辈子也在以家庭作为其电影的不变主题. 父与女, 母与女, 姐与妹的各种不同组合, 小津找到了一种视觉的美学与及剧本的语境. 凝结成一种独有的风格. 杜琪峰有这种创作上的寻找, 杜的指针人物不是小津是黑泽明, 黑泽明和小津的不一样, 在于黑泽题才上的多元, 与及形式的不段求变, 但黑泽明是形式和叙事的高手, 每部电影都在寻找一种形式来表达其剧情, 这是社琪峰期望达到的境界, 也是众多华人导演常常视黑泽明为榜样的理由. 但华人导演总是犯了一个只看见黑泽明的形式, 而忽视了黑泽明的人本追求, 对人性善与恶的分析, 对人生的看法. 黑泽明都是没有放弃过的, 形式只是一种表达的工具.

华人男性导演常常都有认命情意结, 尤其期望名利双收的大导, 看透华人电影业那种冷漠与反人本, 在其作品都在表露对人性失落的潜意识, 张艺谋也好, 王家卫也好, 都有着这种态度. 杜琪峰有点不一样, 但总是逃不过这种悲观的潜意识. 黑社会系列电影正正是社琪峰对香港政冶的一次悲呜, 但也反映了社对中港政冶的简单权力主客定位. 其风格化的场面设计, 只是把一些典型港式恐共情意结重演一次, 对民主失落的反射.最可惜的是没有反思中港权力互动之下, 香港应负上什么责任和可以办演的角色, 古天乐的角色是在无可选择下的被迫上大佬之位, 但这种被迫只是一种假象, 黑道的游戏也和政治一样是一种权力的争斗, 不可能置身事外, 中国人对民主的看法, 从来都带有黑道式民主的味道, 黑道中的义只是一种权力手段, 而不是理想.而中国式民主重视的是个人绝对利益, 而不是人本的正义. 港人在争取民主的历史也只着垂一种一人一票的形式主义民主, 而不是人本价值观的建立.

黑社会电影系列只是把不能实行民主的责任, 推在一个所谓”中国”的身上. 并没有反省”中国”的行为也是香港人行为的反射. 港式投机文化独大完全是一种港人自我放弃自甘隋落认命的表现.

dbanotes posted a photo:

这家伙来了快一年了。胆子特别小。刚来的时候很瘦，现在胖的圆滚滚。和猫泽西关系不错，不打架。缺点是屎很臭。每次收拾猫砂都能把我熏个半死。

晚上参加一当地乡镇企业家的生日宴会，晚上发了张票，去木府看音乐歌舞剧演出《鲁班鲁饶》。鲁班鲁饶是纳西话，意思是只有青春和爱的国度。传说中，这个地方位于玉龙雪山的某地。纳西族因为殉情而死的青年男女都会去到那里，永享青春和爱情。本来想去看一场，但是投资人居然是李亚鹏，那就算了。不知道导演是怎么想的，《鲁班鲁饶》是纳西话。如果当年的《孔雀公主》改名叫《纳木露娜》，不知道结局会是怎样？

刚好有朋友打电话来，又到了一月一次的聚会，我吃完了第一顿又去赶第二场。虽然是云南最常见的天麻炖鸡，但是鸡是土鸡，而且用手撕碎，味道很是不错。吃完饭去2416酒吧，刚好赶上万圣节。在院子里烤火听地下乐队演出，还算不错。最近已经把据点定在了2416，喜欢看各种莫名其妙的艺术家表演。

由于受不了我的打击，饭桌上就预订了30只大闸蟹，一条石斑鱼，还有象拔蚌若干，约了明天去吃海鲜。其实，我是个很厚道的人，并没有要求吃一米五的苏眉。明天飞机从广州空运过来，也可以让我得偿夙愿，和螃蟹共舞。

万圣节，Blog里的傻鸟数量明显减少，值得庆贺。

现在常常碰到这样的事情，浏览网页，看到一个链接，正准备点过去，鼠标移到链接位置就好像被什么东西沾住一样，然后拖拖拉拉嘎吱嘎吱弹出一个比邮票大一点的tooltip，让你预览一下这个链接对应的地址。

这真是一种很差的用户体验。

一个网页，又不是什么两个小时的电影，我还要看个预告片再决定点不点过去。再说了，你那个屁大点的预览，除了提供这种可怕服务的网站的logo（一般都叫snap什么），基本什么也看不清楚了。至少，你也应该做到象普通tooltip一样的延迟加载，别那么热情的就蹦出来。

强烈建议大家不要搞这种中看不中用的东西。

1.      PQO make sense universally on any SMP or multi-processer database server 

Not really. Why use PQO on a small dual core server just because it’s available? And why use PQO on a big SMP server with lots of CPU’s, but maybe not as robust an IO subsystem – what advantage does that offer? And my favorite abuse, we have a 64 CPU server, so we’ll just define all important tables and indexes as parallel 64 (or even some lesser number significantly greater than 1). How many of those peoples’ servers will be running many concurrent queries anyhow? So why sub-divide the work and force sharing CPU’s across processes? We don’t drive the posted speed limit simply because we can if the weather makes that a questionable choice. So we should use the same discretion when planning PQO usage.

2.      PQO “by definition” will make any large operation faster when “parallelized” 

Right – and I have some wonderful “magic beans” to sell to people who blindly believe that theory. Oracle has to spawn additional shared server processes (or reuse those that are lying around still from recent completed PQO operations). So the cost of spawning additional processes and their footprints must be accounted for in the grand scheme of things. Plus, Oracle will need a “coordinator” process to coalesce all the results – and that too has a small cost. So calculate all this into your assumptions before using PQO. But for those of us less scientific, think of PQO as being like a grocery store checkout line. If you only have a couple items in your grocery cart, will forcibly spreading them across multiple checkout lanes get you out the any door sooner? And even if you do have an overly full cart, will spreading your items across checkout lanes that are already busy be any better? There are obvious times (i.e. rules) that are better than others. We simply need to apply the same clear thinking when using PQO.

3.      When I say parallel N, Oracle only exactly and only spawns N server processes. 

OK – so now I also have the clear rights to sell the “Brooklyn Bridge” to people who may think this. Oracle and parallel queries are much more complicated than this. For example an ORDER BY or GROUP BY will double the number of non-coordinator processes required. So we have at least two times N now. But then there are “init.ora” parameters to limit or govern PQO usage. So some queries might get the full PQO amount, some might execute linearly, while others may use <= N processes due to such configuration parameters. The only way to know for sure is to monitor it live.

4.      I have a super-fast IO subsystem and multiple top-end CPU’s, so I can increase PQO usage a “wee bit” beyond my actual CPU count.

OK – I’m 100% guilty of doing this on occasion. In fact, I might even do this more often than not. But I am cognoscente of the above issues and have planned according – and still have a reasonable PQO scenario where my IO is 100% not my chief bottleneck. Thus in those clear cases I will extend beyond the actual CPU count in my PQO settings. I’ve even gone two to four times higher in the right circumstances. And in those cases it has made a clear and measureable difference. But be careful of when it does not make sense. For an easy test to verify this, simply set PQO to degree four on your dual core desktop or laptop and run several concurrent queries. It will not only run slower, but painfully so.

5.      I have a big multi-processer server, so full-table scans are always the way to go. 

As silly as this may sound at first, I’ve seen this in many data warehousing scenarios. I think it happens because somewhere someone wrote that “parallel full-table scans were efficient”. But that author meant in clearly reasonable scenarios – and not universally. Yet this remains one of the top data warehousing tuning mistakes I see. Think about a DW query – it may scan millions of rows in tables with many billions of rows. Why force a full-table scan when there is at least a ten-to-one, if not a one-hundred-to-one, reduction factor. That’s exactly what indexes are for – and index scans can be done in parallel. I’ve seen improvements as large as 1000% from undoing this simple PQO misuse.

Thanks to everyone who submitted an entry to the pumpkin contest!

Matt Lind took us up on our suggestion of a werewolf, and we’re awarding him the $50 for it. He even carved a Fedora logo into the werewolf’s face and put a Red Hat fedora on top.

Red Hatter Brian Kearney wasn’t eligible for the prize, but his family carved up this great trio: Shadowman, a traditional jack-o-lantern, and his son’s Harry Potter inspired “zap” pumpkin.

Finally, we really enjoyed this digital art submission from Mola Pahnadayan.

Thanks again, everyone!

Technological advances continue to amaze me.

I remember setting up my first token ring network.  Huge cables, not very user friendly configuration (why would everyone want to run a cable to a network box like that), everything really big, really slow.

That was almost twenty years ago.

Today, we see things like this. An SD card with enough hardware to be a wi-fi device and store 2gb of stuff.  Or, if you are cheap like me - a 1gb normal SD card for $10 (yes, I had to buy it just because - it was $10)  Now I have lots of room for pictures.

Things are moving scary fast sometimes.  Hard to keep up with everything.

About 10 years ago (ok, 10 years and a month ago), I was asked to review a new PC configuration for a friend, this is the document they sent me from the vendor:

This machine would take about 2 weeks to deliver from the date of purchase.
Gateway 2000 G6-233

Basic Machine
• Processor: Intel 233MHz Pentium II Processor w/ MMX Technology
• Memory: 64MB SDRAM
• Cache: Internal 512K L2 secondary write-back cache
• Monitor: CrystalScan700 17" color monitor
• Graphics Accelerator: nVidia 4MB AGP Graphics Accelerator
• Hard Drive: 4.0GB Ultra ATA hard drive
• Floppy Drive: 3.5" 1.44MB diskette drive
• CD-ROM: 12X min./24X max. CD-ROM drive
• Multimedia Package: Ensoniq wavetable sound card & Boston Acoustics Micromedia Speaker System
• Fax/Modem: TelePath 33.6 Data / 14.4 Fax Modem
• Case: G-Series Mid Tower
• Keyboard: 104+ Keyboard
• Mouse: MS IntelliPoint Mouse

Bundled Software

• MS Money 97 & Quicken SE
• MMX Technology Bundle
• Microsoft Windows 95
• Application Software: MS Office 97, Professional Edition

Service
• Service Program: Gateway Gold Premium Service and Support (3yrs. Onsite) added: US$99

Additional Peripheral Hardware
This hardware is highly recommended.

The scanner, in conjunction with the fax/modem above, replaces the need for a separate fax machine (to fax outbound, you would scan and ‘print’, to receive faxes, the computer would just be turned on). Additionally, the scanner allows you to digitally store press releases, magazine articles, photographs an so on.

The digital camera allows you to capture electronically what it not printed in the papers.

The tape backup unit is a must if you are going to store items you wish to keep for a lifetime. The hard disk will fail at some point. If you go to the trouble of scanning and storing keepsake items, you do not want to lose them in a hard disk failure. The tape device is large enough to backup the entire system overnight, an operation you would want to do once a week/every other week.

We investigated color LCD monitors (space savings, an LCD monitor would be about 2 inches thick, the CRT above is about 17 inches deep) but feel that at this time they are still too new and not readily in supply. The cost of the LCD screen would be as much as the base system itself ($3,500-$4,000 for a 14 inch LCD display). We are not recommending this at this time.

• Scanner: Hewlett-Packard ScanJet 5Pse Scanner
• Digital Camera: Epson Photo PC 500
• Tape Backup Unit: TR4 SCSI TBU and tape (w/SCSI controller)

Pricing
Total Price: US $4200

What cost $4200 in 1997 would cost $5218.48 in 2006.  10 years ago - state of the art was 64mb of RAM, 4gb hard disk.  And cost five times what you would expect to pay today.

Nothing to add to that, just find it amazing sometimes.

　　这套纸笔的GTD应用系统，尽量追求简单，方便，是弥缝在参考了几个GTD Moleskine Hacks后，与自己已有的一些经验，结合精简而成的。这篇介绍将很详细，但并非每一个步骤都对你是必要的，所以大家应该再按照自己的需求进行改进或者进一步地简化。

需要的材料：

• 一个硬壳厚本的、质量好的笔记本，页数在200～300页之间，外带一条书签带子的。弥缝推荐过的理想的笔记本Moleskine，就是很不错的本子。品质优良、设计独到、还有悠长的历史，不过价格就是相对比较贵。
• 标签（Tabs），需要多种颜色的，可以自由移动的。
• 一支好笔，好本配好笔，不仅仅看起来漂亮，写起来也顺畅。
• 了解什么是GTD，你可以阅读《尽管去做——无压工作的艺术》或者阅读Zen To Done来获取这套最受欢迎的时间管理知识。

规划你的笔记本：

• 标上页码。这是件麻烦的事情，想偷懒就可以不做（我就没有做）。下面使用的是一本200页的笔记本为例，你知道在大概相同的页数贴上相应的标签就可以了。
  

  

• 创意的第一页。写上自己名字、随意点缀一下、写些名言警句等等。随便你用来干什么，你可以写上自己的联系地址，如果你丢失了，说不定有好心人会归还给你；你可以写一篇《陋“书”铭》来秀秀自己的文字；还可以是一张规划这本Moleskine的思维导图。当然你也可以不要，这是弥缝自己的喜好而已。
• 在第四页贴上第一个标签，写上“收件箱/任务清单”。你应该已经明白收件箱和任务清单的区别了。收件箱只是简单地将信息全部收集起来，而任务清单中则是对收件箱内信息进行处理后留下的可执行的行动。弥缝把第四页用来收集，第五页用来放任务清单（方便相互查看）。当这两页写满以后，就开始下两页。同时也把标签也移动到新的下一页，而且上一份任务清单中未处理完的任务也要存到新的任务清单中来。
  

  

• 在第102页贴上第二个标签，写上“项目清单”。当一件任务需要多个行动才能完成，那这件任务就是一个项目，那么它就应该放到“项目清单”中。每一页只记录一个项目，你可以把书签带子夹在现在正在进行的项目页面上。
• 在第180页贴上第三个标签，写上“Someday/Maybe清单”。这里用来记录某一天可能去完成的事情，可能也是一个理想或者长远的计划等等。
• 在最后一页贴上一份全年日历。Moleskine的背面还有一个Pocket，你可以用来存放一些发票、随意贴等等。
  

  

使用固定记录格式：

　　记录的格式因人而异，每个人都有自己的喜欢的记录方式。只要你觉得适合自己就可以了。通常来讲，你的格式中应该包括这么些内容：时间、任务、地点。这里弥缝只是简单地演示一种普遍使用的列表记录格式：

[时间] [拖延次数] [任务] [地点] [任务/项目]

　　比方说你想去买一台新笔记本电脑，你就可以在任务清单中这么记录：

10月30日 (0) 买笔记本电脑 电脑大市场

　　当然，一分析你就会发现上面这一件事并不是单一行动的，而是一个项目。所有你就应该把它归纳到“项目清单”中去。你应该为每一个新的项目都建立新的一页“项目清单”，并且在这一页的开头写下这是什么项目，比如说：

项目：购买笔记本电脑

10月28日 (0) 比较笔记本评测 网络

10月29日 (0) 到电脑大市场去调查价格 电脑大市场

10月30日 (0) 选定机型，付钱购买 电脑大市场

　　为了便于区分，你还可以使用不同的颜色的笔来记录，比如说用红色来标明时间、橙色来标明拖延次数、用绿色标明地点等等。

　　当你完成任何一件任务时，要记得从清单上划除。如果你在开始新的一页任务清单时，旧的任务清单上还有未完成的任务，你就要把它记录到新的任务清单上，并且在拖延次数上加一。如果某一件任务拖延次数达到三或者更多时，你就该赶快完成或者考虑是否还需要去完成了。

　　最后提醒，随身携带你的笔记本，并且及时把信息记录到收件箱中，而且也要信息及时处理掉，信息处理后也要划掉以免返工。对于信息的处理参考弥缝以前写的《ThinkingRock：最好的GTD软件2——使用方法》中介绍的方法。

　　如果你有疑问，欢迎提问。这套系统可能有不完善的地方，你若有不同一般的却非常有效的方法，欢迎分享。

本文为转载文章，原文于2007年10月30日发表在褪墨上。
弥缝授权广善对此文进行转载。

©作者：Fenng 发布在 dbanotes.net

WordPress.com 母公司 Automattic刚收购 Gravatar 没几天，工程师就对 Gravatar 进行了一番手术，把 Gravatar 并入了 WordPress.com 的技术架构.

合并后的 Gravatar 在两个不同的数据中心各有一台应用服务器 + 1 台 Cache 服务器。Cache 服务器用的 软件事 Varnish ，峰值能够处理 1000个/秒 的请求，效率很惊人，据说 Varnish 跑在 FreeBSD 6 或是Linux 2.6 上充分发挥性能，实际处理能力比这个还要强。

Web 服务器分两种：普通的为 Apache2 + Mongrel, 图片服务器则是 lighttpd + mod_magnet (看来 lighttpd 是图片服务器非常流行的使用啦 )，不过他们遇到了内存泄漏问题(Bug?)，每隔一段时间要重新启动一次，对这个的控制用的是 Monit

Monit 这个小工具我是第一次知道，功能也很有趣。

小成本，高性能，这帮老外玩的就是透。国内的 Feedsky 啥的也需要加把劲儿了，最起码也要向豆瓣看齐吧?

--EOF--

相关文章|Related Articles

• Slashdot 网站架构补遗
• Amazon 的 Dynamo 架构
• Digg 网站架构
• 《大型 Web2.0 站点构建技术初探 》一文系抄袭

评论数量(3)|Add Comments

本文网址：http://www.dbanotes.net/review/wordpress_gravatar_arch.html

作者：雅虎博客搜索产品组

为什么用博客搜索？

有个朋友十一独自去青海，地图没有带，带了一份攻略。当她把攻略分享给偶遇的“驴友”时，别人很奇怪：为什么他们找不到那么好的攻略？她很得意，告诉大家她是用博客搜索找到的。

在博客里，有着成千上万的 “意见领袖”，他们无私地贡献他们的智慧，分享着各种经验，发表着各种新鲜观点。其中不乏具有相当影响力的博主，让众多的用户有意外的收获。“领袖”周围总是会聚集很多人，他们可能有这共同的爱好和兴趣，从这些人中可以看到更多的新鲜话题，认识更多的朋友。

在现实中，一旦出现了新闻，一定会有无数博主第一时间在博客中发表评论，让我们看到更全面的信息。所以用博客搜索可以帮助我们找到网友的观点和经验，可以找到更多有着相同兴趣的虚拟圈子，有着更多流行话题。

博客有了新分类

现在，播客越来越流行，大家不仅能从播主发布的视频中找到各种视频，同样可以发现很多有着共同兴趣爱好的播主。这也是我们把播客纳入到博客搜索范围的原因。

在博客搜索中增加了新的博客分类：博主，文章，播主和视频。希望大家能更有针对性地找到想要的东西。

排序筛选帮您忙

在许多用户反馈中，希望增加时间的排序。本次版本的升级中，增加了时间加权的排序方式，在相关的前提下让大家看到最新的文章。

您的博客粉丝有多少？

哪一个辛勤耕耘的博主不希望自己的文章受大家欢迎呢？有谁不希望知道到底谁在关注自己的博客呢？当然未公开的博客主除外了。

在雅虎博客搜索里，有一个小功能可以帮助各位博主找到自己的粉丝在哪里。这是通过搜索引擎的网页链接关系，找到引用或指向您的博主有哪些。您可能会惊喜地发现有不少陌生的朋友在关注您。

在这里我们不得不举老徐做例子了：徐静蕾的博客粉丝人数可能有10400。

那么您的呢？希望您也有惊喜。也许在哪些博主中您可以发现共同兴趣和爱好的朋友。

流行度意味什么？

在粉丝团和文章页面大家会发现“流行度”排序。流行度是搜索引擎通过博客和文章的综合表现，包括更新频度、文章数、受欢迎程度等因素给博主和文章打的分数。在文章列表和粉丝团列表里，您可以方便知道您的哪些文章和哪些可能的“粉丝们”更受搜索用户的欢迎。

博客没有收录怎么办？

如果博客没有收录，不用着急，您可以自主提交RSS和博客地址，搜索引擎会自动索引和展现您的博客。

热门话题的自动推荐

经常有朋友提到，我不知道在博客里搜什么？

其实，在博客文章里包含了很多的社会话题和热点讨论。搜索引擎是可以通过用户搜索量、变化量等等因素的综合，推送出来网友正在热议的各种话题和受欢迎的博主。

在全能搜索的博客主页中，推荐的有播客视频、博客文章，内容包含了各种分类，大家可以对网络中流行的话题和视频了如指掌。

如下图：

期待您使用博客搜索，让网友的智慧帮助您。

    Few days ago, somebody (they wrote a MYjDUL utility) post a message on a public forum said that AUL/MyDUL is copied from jDUL open source project. This is not true, when I started to write AUL/MyDUL utility in 2004 Q4, I did not know this open source project at all, else I may not start to write this utility. The first version is also written in Java language, I spent a lot of time in AnySQL utility with Java programming, so Java is the language I know best at that time.

    You can download the source code here, I redesigned and rewrote it with standard C language for self-protection and for high performance. There are three major files.

1, OracleBlock.java -- Interface Defination
2, OracleBlockV2Big.java -- Implemention for Big Edian
3, OracleBlockV2Little.java -- Implemention for Little Edian

    But I don't remember how to get it work, but you can still get to know the Oracle data block format from the source code lines.

    Although it's written in Java, but not a jDUL copy.

Related Posts

Leave New Comment(Current: 0)

Link: http://www.dbatools.net/mytools/mydul_java_source_code.html