10/28/2012

地理与信息

今天读到两则有趣的报导。一则强调地理在当今信息时代重新显示出其重要性 (A sense of place, http://www.economist.com/news/special-report/21565007-geography-matters-much-ever-despite-digital-revolution-says-patrick-lane),另一则则表示地理优势在信息经济新时代已经越来越不明显 (亚马逊VS沃尔玛:信息经济时代零售霸主逆战,http://www.21cbr.com/html/topic/201210/28-10535.html)。

第一篇文章回顾了历史上对地理的三种看法。第一种是出现在90年代的end of geography。由于互联网的共享性,全世界任何一个角落的人理论上都能享受到共同的资源。在网络上不再有距离的概念(the death of distance)。第二种概念更加极端,认为网上的虚拟世界可以完全替代真实世界。我想这是在21世纪初随着SecondLife这样的cyberspace兴起时的思潮。在这样的虚拟世界里用户可以模拟现实生活中的几乎一切事情,甚至有自己的货币。然而这种想法被证明是不现实的,虚拟世界不可能脱离现实存在,比如虚拟的货币仍需要通过真实货币来兑换。
第三种认为现实世界会影响网上的行为。正如地学第一定律所述(Everything is related and things closer in space are more related),用户更关注身边的事 (local)。随着移动设备的普及和地理定位技术的成熟,企业提供这种local service成为可能。从这个角度讲,geography成为local business越来越重要的因素。

第二篇文章比较了亚马逊和沃尔玛两个零售业巨人。后者是传统的线下销售,前者是这个信息时代特有的产物。近年来亚马逊的营业额逐年上涨,剑指行业老大沃尔玛。沃尔玛通过数十年努力建成的由土地、建筑构成的实体网络正在遭遇前所未有的危机。网络商店的出现,让传统零售业的法宝 location, location, location显得不再那么重要。而目前,两家公司都采取着相似的战略:线上线下的有机结合。

曾经有人疑惑,网络到底需不需要有距离的概念?这涉及到一个问题,即地理到底应该在多大程度上与网络、与信息结合?这里面涉及一系列问题,包括人对(地理)信息的需求,人感知(地理)信息的偏好和习惯,(地理)信息系统如何满足人的认知习惯等等。作为地理信息科学的研究人员,我觉得我们的时代马上到来了。

10/13/2012

Academia is not the only option for PhD

It has been puzzling me all the time, even after I've come here for PhD, that how is PhD accepted in industry. There are many negative sayings about PhD: "Permanent Head Damaged", because PhDs are usually stubborn and confined in a narrow field; old aged, and often immersed in old fashioned papers published decades ago; idealist, whose theories and ideas are built in the lack of technological support; etc. So it's exciting that today there is a doctoral career exploration workshop, with most of the speakers graduating from doctoral programs of Penn State. And I should say, I really learn a lot. Or to be more exact, I actually knew those stuff. I simply got lost after I dived into the mess. And the speech today helped me out, and reminded me of my strength as a PhD.

So why does a company hire a PhD. Apparently the company has a different expectation for PhD. The company won't pay a PhD to program. Some of the things that a PhD is expected might be:

1. Direct a team. Think of ideas in what directions should your team go. Be creative and innovative.
2. Theorize your product. Apply the theories in your domain to practice. It's important that your products are well theoretically founded, so that it is convincing for investors and customers. Be rational and persuasive.

We should be aware that the graduation program is not to get us damaged, but to get us fully prepared.
Keep on asking ourselves internally what is my goal, what is my value, what is the need of the world. Find a joint point of the three aspects, and we can make a difference.

For me, I actually already have a very clear path. I'm to study the cognitive aspect of computer systems. The problem for me is what I can do to improve the usability of a system. And I would start from how people perceive information, and apply that understanding to the design of systems. As I have a GIS background, I may focus on GIS as the study case, but I should also not limit myself in GIS. Therefore my strength should definitely not be interface design or system programing, but to provide guidelines to those guys, in a theoretical level, and suggest evaluation methods before the final product is issued to the public.

So what I need to do in the first step is to enrich myself with the theories of human cognition. Also to leverage my strength, I should apply those theories to the GIS field. So a deeper understanding of GIS users and their thinking, reasoning, and behavior is required.

Finally, building a friend network is as important as academic achievement. I should not lock myself in the lab all day long any more. I should get out, talk to people in other areas deeply, and get inspiration.


9/03/2012

Information Science

During the overwhelming first week, I have been enriched with many new conceptions or familiar conceptions that didn't make so much sense to me. So many that I don't know which to begin with. Maybe, the name of my department, IST, is a good start.

Founded in 1999, IST (Information Sciences and Technology) is a comparatively new college. It is about Information, Technology, and People. There used to be a debate about the third element: should it be People or Users, users of information and technology? Finally People wins, because proponents argue that we not only care about users who use our technology, but also those who don't. To analyze why they don't use our technology helps us keep moving too. I think that's a good idea, and the idea lays the foundation that IST is a discipline centering on humans.

So what is information science? And what is information?

These seem to be very broad problems, problems we generally never come down to think about. We keep on talking about "information", "information age", "information science" everyday, but we actually take them for granted and never give a second thought about they are "intentionally".

Personally I have been immersed in GIS for four years. Shamefully I now find my understanding of it far from enough. I have always been thinking of it as a system, or a science of geographic information. And more accurately, it's just about geographic data. Open an arbitrary GIS textbook, and you'll find all it deals with is data: data acquisition, data storage, data analysis, and data presentation, although some may touch a bit on the difference between data and information in the beginning of the book. Now I come to realize that GIS can also be viewed as a subset of "information science", in the geographic domain. This gives us broader views of GIS, and can be guided by some more mature theories in information science.

Of course, just like GIS, as a newly emerging discipline (emerged in the 1950s), there are still many disputions over concepts, definitions and theories in information science. The difference is, information science has been drawing wide attention both from government and society from the very beginning, and its wide application appeals to scholars in various disciplines, which guarantee its fast booming over the last few decades, both theoretically and pragmatically.

Information science is driven by problems. To solve a problem, we need information. We seek relative data akin to problem, and organize the data into information trying to answer what, when, where, and who. Finally we come to conclusion "why" this problem happens and "how" to solve the problem, which ends up in the so-called "knowledge". This is the basic data-information-knowledge-wisdom hierarchy proposed by Ackoff 1989.

Technology advances rapidly, so we are faced with ever increasingly large amounts of data and ever increasingly complex systems to automatically deal with these data. But no matter whether we are generating data or consuming data, one thing we have to keep in mind is that our single task is to solve problem. We are not developing awesome software, but designing systems that better assis people to solve a problem. In this sense, "people" should always be the center of research. We try to understand how people learn and use information.

There are three major branches in information science: information retrieval, information relevance, and information interaction. Information retrieval is the basic need in this information overload age. And information relevance is closely related to the effectiveness of retrieval. They are more concerned with algorithms, and user participance is limited, with only a query input. In contrast, information interaction is, in my opinion, more of upper level. It asks the system and the user to work together in an interactive way, to co-solve a problem. It is what I will be engaged in, I believe, the so-called human-computer interaction, or more specifically, human-GIS interaction.

The overwhelming first week

开学第一周,相当慌乱。

共选了五门课,其中三门大课,一门IST的关于怎么阅读文献的,一门IST的information management,还有一门Geography的Geovisual analytics Seminar。每门课都要做大量的文献阅读。

现在才知道原来课可以这么上:不需要你有太多的基础(prior knowledge),而是通过阅读来快速学习,然后通过课堂的分享和讨论来培养critical thinking。想起蔡老师跟我讲的,作为一个phd,你可以一开始不懂某个领域; 但一个星期以后,你不仅要了解,还要能说出个1234来。

这里的学生可能就是这种教育的成果。They keep talking。而这正是我缺乏的:怎么把输入整合之后输出并传播出去?

第一周遇到的很大的问题是阅读文献的速度。四篇文献约100页,我花了两整天才读完(当然也包括经常读不下去倒头睡觉的时间)。以后要注意阅读的方法。The Thinker's Guide to Analytic Thinking中提到的关于reasoning的几个要素,可以借鉴下:
另一个遇到的问题是,我完全听不懂学生的发言。学生的发言与老师的讲话不同,他们极其随意、含糊,而且语速很快、没有停顿。而如果听不懂他们的发言我就很难参与课堂讨论。

在Geovisual Analytics的课上终于见到了传说中的Alan MacEachren,看上去年纪很大了,可是却非常健硕,眼神非常犀利。据说他的课任务非常重,从他的syllabus可见一斑:课程的project希望能至少放会议上发表。Sigh...

不过第一周还是相当充实的,接触了很多新的概念,有一种迅速膨胀的感觉。接下来,加油!

8/08/2012

来美第二天

在陈旭师兄家的沙发中胡乱睡了一觉,没倒过时差来,只睡了两个多小时。

早上8点出发去学校了。这里的公交车系统算是美国比较发达的了,可以去小镇的好多地方,但因为是暑假,班次比较少,约20-30分钟一班。

PSU学校很大,可是并不让人惊艳。建筑都是中规中矩的,既没有西方传统的高大宏伟,也没有历史沉淀的古典优雅。大部都是砖红色的,像内地新建的高中。不过校园内高大的树木和广袤的草坪给人很宁静的感觉,还有随处可见的松鼠,让人忍俊不禁。

先去了Boucke Building的Global Program处做了document checking,然后在hub的Id+这个办公室办了学生证和PNC银行卡。两者可以绑定,非常方便。之后是web access account的激活,差不多报到手续就结束了。系里的报到得等弄完SSN(social security number)后填完I-9表格(表示有受佣资格,可以做assistant)后才可以,干脆放在orientation时做。

中午在hub里的panda express吃的中餐,点了西兰花和牛肉,味道不错。

下午去了学校最大的图书馆Pattee Library,有点小失望。虽然藏书应该很丰富,但布置毫无特色,像内地的图书馆,在书架边上放着一列大桌子供人学习,没有典雅的台灯,没有励志的壮语。

晚上到了传说中的parkway plaza,房间旧了一点,但还挺宽敞,室友人也不错,发现可以住,就当晚托学长把东西运过来了。师兄给了我一张单人床(matress + box),碰巧另一个师兄要搬走,去他那淘了不少东西,晚上两辆车浩浩荡荡地就开过来了。好玩的是,床垫和下面的box是放在车顶运过来的,我坐在车厢里用手扶着,感觉就像视频里开挂了的印度卡车一样。

终于有了自己的住所,以后像家一样好好经营。

8/07/2012

来美第一天

北京时间2012年8月6日6点12分,在贵博的寝室醒来,开始洗漱。7点,坐上第一班经过联合苑开往大学站的小巴。搭MTR到沙田,坐上直达机场的通天巴士(air bus,中文翻译更霸气)A41。8点20分,到达香港机场 terminal 1。

Delta的值机柜台在D区,排了一条长长的队伍。称行李时发现超重了,27Kg,比规定多了4Kg,无奈只好交了580港币,后来一想其实可以把箱子里的小书包和衣物拿出来,因为看很多人其实拿了两三件小包登机。

9点在机场吃了大快活。可能最后一次吃大快活的早餐了,点的滑蛋炸鱼柳,竟然感觉味道不错。

9点20进入安检、出关,9点40到达31号闸口。不一会儿就登机了。

从香港直飞底特律(Detroit),历经约15个小时。飞机上的电影大部分竟没有字幕,发现自己还真听不懂对白,在美国有的受了。

美国东部时间下午1点20左右,飞机在Detroit降落。本来想把沿途拍下来给亲爱的作参考,后来发现指示牌非常清楚,又有工作人员指导,肯定不会走错,惟一需要注意的是入关有两个通道:US citizens和foreign visitors,我们要属于后者。拿了行李,开始入关,将护照 、I-20和飞机上填的I-94给海关人员,他会在I-20上盖章,并把I-94的出境卡钉到护照上,这两个文件非常重要,要小心保管。

走两步是入关物品检查,会有只警犬闻你的行李,警犬竟不是那种大狼狗,而是那种特可爱的beable(听后面的美国小男孩说的)。海关会问你带了多少现金,要求不多于10000美金。

海关的人其实并不严肃,会和你聊天。当我说要去Penn State时,他长长地叹了一口气:Oh boy。原来是最近发生的橄榄球虐童事件,看来给学校的声誉造成了很大的影响。

整个转机手续用了约40分钟就完成了,因此之前担心的转机时间不够(2小时)根本不是问题。2点多就坐在B6闸口等去State College的飞机。可是却迟迟不来。据说是打印不出乘客的信息?Paperwork之类的,听不太明白。不过那个检票的大叔态度挺好,很详细地向我们解释晚点的原因。

终于在5点40登机了。飞机真的是非常小,从闸口到飞机没有以往的通道,只用简单的一个木板架住,我的登山包放不进飞机的行李箱,只好拿去托运。起飞没多久我就沉沉地睡着了,醒来时竟然已经着陆了,没能鸟瞰一下整个State College。

下飞机时享受了国家领导人的待遇,从机舱沿着台阶走下来的,四周非常空旷,虽然已经下午7点10分,但太阳仍然高照,天空蓝的出奇,我一下子被这种乡村的空旷的宁静的美丽震憾了。

拿了行李没多久,于博师兄就来了。路上得知他的老婆和妈妈都在,马上就有baby了,在美国有个温馨的house,让我马上憧憬起我和瓜瓜的未来。晚上在麦当劳随便吃的快餐,6刀多。然后晚上和他家里人一起去了Walmat,买了点生活用品。那个超市很大,还卖家具,24小时营业。神奇的是收银台没有人,是自助check out的。将物品的条形码一刷就好。有趣的是,你还必须在刷完条码后把这件东西放入旁边的购物袋内,否则是不能刷第二件物品的。

这几天暂住一个叫陈旭的师兄这,在copper beach W. Aaron Dr.。睡沙发,竟然睡不着,12点躺下竟然2点半就醒了。折腾到3点半,索性起来,记下来美国的第一天。

第二天,也就是今天,去学校办各种证件。看看这个我未来要呆五六年的地方长什么样。

6/29/2012

GeoInformatics 2012

2012年6月15日至17日,GeoInformatics 2012在香港中文大学举办,由我们实验室承办。GeoInformatics是CPGIS这个华人地理信息协会每年都要举办的国际性会议。1992年,在美国纽约Buffalo大学,几个年轻的留美博士生自发组织成立了这样一个华人协会,领头的就是现在的林珲教授。20年后,当时的协会成员或成为工业界的领头人,或成为国内外顶尖大学的学者、教授,共同支撑着GIS学科的发展。我能想像这种从无到有的成就感,也很感叹这帮人敢想敢干的精神。

这届的会议非常成功,也邀请到了许多国际GIS大佬,如Goodchild, Michael Batty等。甚至还请来了GIS之父Tomlinson. 当2米多高的Tomlinson拄着拐杖颤颤巍巍地走上讲台,以一个父辈的身份,用颤抖的、接近哭腔的口吻对我们说“Now, I talk to you directly, the future of GIS”时,我仿佛接受了一次洗礼,似乎那个学科发展的巨担已传递到了我们肩头。

会上我还意外地碰到了John Radke,让我一下子想起了三年前的我。那会儿我利用寒假的时间去美国UC Davis交流,期间去了UC Berkeley. 因时间仓促,事先没联系好那边的老师,只在网上查了地理系的地址,就凭着一股年轻人的蛮不讲理去了。在系楼的过道上看到了John Radke的名字,只知道是和GIS相关的,就和秘书说我要找他。没想到他并没有怪我来得莽撞。很奇怪的,我们并没有讨论GIS的话题,而是就中国的问题争论起来。他说他小时候在中国长大,对中国很失望,中国的政府和学术很腐败;我便和他说现在已经不一样了,你大可以再去看一下。争论的最后我们都按捺不住脾气了,一小时的谈话就终止了。现在想来,当时的我真是好笑,而John的风度也的确让我印象深刻:一个大牌教授突然被一个不知哪来的毛头小孩打扰,还莫名其妙地和他讨论了一小时中国问题!记得讨论完的我异常兴奋,回去后还给他发了邮件,贴在下面,以供玩味。
Prof. Radke,


     Hello, I'm the Chinese student who met you this afternoon. I'm from Zhejiang University with GIS as my major.
I really admire you, not for your great achievements in GIS, but that you are truly thinking about what we are doing and what we are doing that for. You made me think deep into why I learn GIS. I have too much to say, so it might be a little bit long letter. 

     Money is necessity for life, but I never worry too much about it. I have my dream which means more to me and I keep it all the way I grow up: I want to do something to change the world( big dream, err?) I keep on asking myself: but how? until I met GIS. GIS tools are powerful but  they are not interface friendly and are complicated to use. Why not make it as popular as photoshop, thus pushing GIS research one step forward?

    But after the talk with you today, something new came into my mind. GIS is a tool, and only a tool which can be used in any way or not in any way at all if you are not determined to solve a problem! There exist many problems--both in developing and developed countries--environmental degradation, resource shortage, transportation problem, etc. Actually the government could solve them, but it costs. What we might do is, to make the cost smallest so that the governor is willing to take the risk, with the tool of GIS. And I might turn to politician if necessary.

     I don't know if it is another naive idea. It just hits my mind on my way back home.

    As for the "China" you mentioned today, I just feel sorry. But it's not apology. I just feel sad that you didn't see the other side of the country. That's not real China. There are many problems in China like government corruption and environmental degradation. But we have identified the problems and we have taken action. We just need time. Time is indeed impressing but it's never too late to take action. China is no longer the poor country before. It is a big responsible country, just look at 2008 Olympics and the important role China plays in the Copenhagen conference. I truly   hope you pay a visit to China once more, and perhaps to Zhejiang University, Hangzhou.

     By the way, Beijing is the capital of P.R.China, while Capital Nanjing is history. They are two different periods of China of 5000 years old.
John竟也马上给我回了一封长信:
Dong Chen,


It was good to meet you.  If I could only transfer my experience to you
directly you would see that in my mind I see the beauty of your home
country and in fact all the counties of the world.  I also see the future
and the custodians of the future.  This is my dilemma. It starts with
humans and their inability to see the bigger more important problems.
They are for the most part goal directed and this is their weakness and
downfall.  No doubt China will fix its current problems but will they
learn?  The US is young, has experienced many growing pains and I believe
rarely learns form those experiences.  If we only had time to go into this
further.


To save a life is to save your own.  Save mother earth and look what you
have done.


Governors (politicians) are usually caught up in themselves and by the
time they realize they could have done something good ... their time is up
and a new one enters the scene.  Faith in their ability to see the
problems and articulate a solution is likely naive but this is not a
failure of yours but theirs.


Spatial reasoning and understanding brings us closer to recognition of
what is.  This is why I have embraced GIS.  I see you feel the same.  I
salute you for it.


kind regards,
这次再见,很可惜他已经不记得我了。希望有一天,在我学有所成之时,能再和他"talk around GIS problems"。

这次会议是我认识GIS圈里人的大好机会,在这里把CPGIS的一些会员贴出来,供膜拜学习(不保证信息完整准确)。

1. 宫鹏。19岁即于南京大学地理系本科毕业,21岁于南京大学地理系硕士毕业,25岁于加拿大滑铁卢大学地理系获得博士学位,现任美国UC Berkeley大学教授,清华大学地球系统科学研究中心主任。
2. 丁跃民。美国Verizon Communication公司信息技术部资深经理。Verizon主要运营无限服务,美国最新的4G网络就是他们运营的。
3. 龚健雅。江西人。毕业于华东地质大学(今东华理工大学)。中国科学院院士,武汉大学教授,武汉大学测绘遥感信息工程国家重点实验室主任。吉奥公司总工。
4. 周成虎。福建人。南京大学陆地水文专业学士,中科院地理所博士。陈述彭学生。中科院地理所副所长,资源与环境信息系统国家重点实验室主任。
5. 李荣兴。上海同济大学surveying and mapping专业学士、硕士,Technical University of Berlin 大学 Photogrammetry and Remote Sensing博士。曾是NASA火星探索计划科学家。现任Director, Mapping & GIS Laboratory,Department of Civil and Environmental Engineering and Geodetic Science, The Ohio State University
6. 李斌。广东人。Ph.D., Syracuse University。现为Central Michigan University地理系教授。
7. 柳林。中山大学地理科学与规划学院院长。美国辛辛那提大学教授。
8. 林戈。美国内布拉斯加大学医学中心副教授。
9. 夏福祥。浙江衢州龙游人。浙江大学毕业。现为ESRI 资深构架师。
10. 周启鸣。香港浸会大学地理系教授,地学计算与分析研究中心主任。
11. 王野乔。美国罗德岛大学自然资源科学系教授。
12. 关蔚禾。哈佛大学地理分析中心研究服务部主任。
13. 涂汉明。Octagon Research公司临床信息技术部总监。
14. 陶阅。湖北人。武汉大学毕业,后赴加拿大卡尔加里大学用两年半时间获得博士学位。32岁获终身教授职称。开发了GlobalView等软件,现为PPLIVE的CEO.