为汶川大地震救灾而建『四川老師』计劃-寻求drupal志愿者开发技术主管人员 Co-head and developers needed for Earthquake Relief Efforts by the "Sichuan Teachers"

(This job might be changed into a paid job in the future if our grant proposals are approved. 如我们获得资助，此工作可能变为受薪职位。)

Sichuan Teachers (https：//match.sichuanteachers.org, http://sichuanteachers.org) is initiated and organized by some volunteers home and abroad who are greatly concerned with the education in the areas affected by the 2008 Sichuan Earthquake. We include professors and students from some of the best universities in the world. We also have professionals in the industries. We gather with a common goal--to provide long-term academic help and tutorials to children directly affected by the 2008 Sichuan Earthquake. 『四川老師』计劃(https//match.sichuanteachers.org, http://sichuanteachers.org)由一群来自欧美，中国等地学生和专业人士自愿发起。当中不乏著名大学的教授、研究生、本科生及中外专业人士。计劃主旨是希望在教育上长期向汶川大地震受灾的学生们提供免费咨询帮助.我们最终希望能從『四川老師』或因受其启发所生的计劃让受灾的每所学校受惠。

Participants are free to choose what they want to teach or learn. The students and volunteers mutually choose each other. This website might use some sophisticated economic mechanisms to match them in an efficient manner. We communicate with the students through real-time Internet services(MSN, QQ), land phones, cell phones, e-mail and postal mail. We could possibly, for example, offer free weekly one-on-one English lessons or trained counseling to students through a sponsored cell phone shared by 5 children, with a student possessing the phone on a school day and pass on to another person the next day. Some of us go to the affected schools in person. We would also like to try our best to look for sponsors to supply communication equipment, although we cannot guarantee that we are able to find one for you in the end. 在教育咨询内容方面，我们目前不会设定教材，在自由选择的原则下，由我们或会用先进的配对系统以高经济效率的方法配对自愿者与其辅导学生,让双方协定教材和辅导内容。因我们大多志愿者身在国外，而灾区学生们在灾后几个月甚至长期所拥有的通讯设备都有限，我们拟采用实时网络通讯(MSN，QQ)，固定电话，移动电话，电子邮件，以及书信方式进行辅导。例如我们可以在每个星期提供一次一对一的英语教学,让五个学生分享一部赞助商提供的手机,轮流接受我们提供的辅导。我们充分认识并理解灾后重建工作的迫切度以及困难度，因此如有需要，我们也会尽力帮助寻找（但不保证能寻获)赞助人及赞助商来解决通讯基础建设所需的资金和设备问题。同时，我们也可能有一部分自愿者愿意親身来到灾区试点辅导，以了解到灾区学生最迫切的教育咨询需求以及长期心理建设需求。

This website stops allowing data entry of personal information from all users on XX/XX. It then might allow users to enter rankings based on personal preferences. The website might display to each side a table of anonymized personal data of the other side. Each side ranks a subset of individuals on the other side. For example, a student sees a list of volunteer teachers. If the list is too long, the student filters the list by listing, say, teachers who teach math first. The student then chooses which ones are his first, second, third choices, etc. Likewise, the teachers also rank the students. This stage lasts for two weeks, and the website is closed temporarily for matching students. 此网页将在XX月XX日停止所有用户输入个人信息并接受用户依据个人倾向排序。此网页或会将为一方显示另一方的匿名个人信息表格。每一方都将对方进行分类排列。例如，学生将会看到志愿老师的列表。如果此列表太长，学生可以通过排序来筛选老师，譬如，学生可以将教数学的老师排在首位。这样学生就可以选择他们自己需要的首要人选，次要人选，再次人选，等等。同样地，老师也将学生进行排列。这项工作将持续约两周，此网页将在XX月XX日暂停并对学生和老师进行配对。

We will reopen this website on XX/XX. We are discussing a few candidates for the matching arrangement and algorithms. One possibility is that our matching team will compute some "socially efficient" scores and match the students and teachers into pairs. The words "socially efficient" are in quotes because it depends on your point of view. For example, considerations that we take into account include stability (i.e., each pair does not tend to break easily across time, which is especially important for orphans), efficiencies (e.g. roughly speaking, we try to give everyone their best choices if feasible). Since 2005 similar algorithms have been implemented into matching students and schools in NYC and Boston. (see, for example, this post). 我们在XX月XX日重新开放此网页。我们已有几个候选配对方案但我们依然在讨论中。一个可能是，配对组将计算一些“社会效应”分值并对学生和老师进行配对。“社会效应”一词在此加了引号，因为这取决于你自己的看法。譬如，我们所考虑的因素包括稳定性（即，每一对老师和学生最好不要轻易解除配对关系，尤其是对孤儿），效率（总的来说，我们希望每个人都得到最适合的配对选择），等。自2005年以来，一些类似算法已在纽约和波士顿等地被运用在学生和学校的配对中（譬如，请见 此文）

The website reopens on XXX or earlier to announce the pairs. The website may reopen a few times a year to allow rematch especially if personal data and preferences change. 网页大概将在X月份甚至更早的时间内重新开放，以公布配对结果。若有需要，一年之内可反复匹配几次，尤其是当个人信息和个人倾向有所改变的时候。

Main Tasks： 主要工作内容：

We need several developers and a co-head to work with our current head of IT group. 目前，“四川老师”项目正需要一些拓展人员，并且IT组还需要一个合作的带头人。

Our matching team already narrowed down some matching mechanisms. Roughly speaking, in terms of informations system needs, see the following. For now, we need you to install your own drupal 5, we will give you our database of questions, then you need to code and implement (and fully tested) #4 and 5 below best by 7/11. It is not too hard if you know drupal well. We will move your code to our system if it works in your system. After that, there may be other tasks if you are interested. Please reply us as soon as possible. 配对组已敲定了一些配对机制。有关信息系统方面的需求，请参考如下内容：目前您需要在您的电脑上安装Drupal 5，我们将为您提供问题的数据库。然后，您需要在7月11号之前进行编码并实施（在完全通过检测的基础上）以下的第4项和第5项。如果您熟知Drupal，那么这并不是一个难题。如果Drupal在您的系统上运转正常，我们将把它转移到我们的系统上来。在此之后，如果您仍有兴趣，可能还有更多工作可做。请您尽快回复我们。

1. Enter application questions in Drupal. (This part is done! We implement using the profile module. A combination of Nodeprofile/Nodefamily/AutoAssignRoll might do a better job. We can avoid using PageRoute. And it can do a better job than Profile.) 在Drupal上输入申请问题。（这部分已完成。我们实施运用profile module。结合Nodeprofile/Nodefamily/AutoAssignRoll也许更好。我们可以避免使用PageRoute。这比profile module更好。）

2. Students enter data electronically. Volunteer teachers enter data likewise. 学生输入电子信息。志愿者老师也输入电子信息。

3. (optional) The verification system or reputation system works to verify the data or compute a reputation score for the data. Do you have a good solution? I feel that a reputation system might be a highly implementable solution. For example, advogato.org allows users to certify each other: http://www.advogato.org/person/claire/ One of the key point is to have a trustworthy set of seed users who start certifying each other. Systems like elance might be too costly and inconvenient for users. （任选项）用核准系统或信誉系统核定输入的信息或计算这些信息的信誉分值。大家有什么好的解决方案吗？我个人认为信誉系统或许是一个相当可行的办法。比如，advogato.org 允许用户互相确认：http://www.advogato.org/person/claire/。其中一个关键点是需要拥有一组可靠的种子用户来开始相互确认。像elance之类的系统可能对用户来说太昂贵且不太方便。

4. Students are presented a list of all teachers (with their data anonymized). Students could choose how to order the list (e.g. by subjects they could offer, by location, etc.). Likewise, teachers also see the list of students. (Optional: Eventually, sponsors could see the list to to select the students, teachers, and the student-teacher pair they want to sponsor.) 学生看到志愿者老师的名单列表（老师的信息是匿名的）。学生可以选择如何对此列表排序（即，根据老师所能教的科目、老师的所在地，等等）。同样如此，志愿者老师也可以看到学生的名单列表。 （任选项：最终，赞助人将会看到名单以选择学生、老师、以及他们愿意赞助的一对或数对“学生-老师”）。

5. Each student ranks all (or a subset) of teachers, and submit to the system. Likewise for teachers. 每个学生对所有老师（或一个子集的老师）进行排序，并提交给系统。老师也如此。

6. The system is closed for say a week. 系统在一周内暂时关闭。

7. The system reopens and displays the all student-teacher pairs matched according to some matching mechanisms. (For the test stage, you don't have to worry about how to implement the mechanisms because our matching design team will do it using matlab, just output the ranking of each student and teacher to the matching team, and you just need to give us a method to input the list of pairs in the system). Note that each student or teacher will have a group ID. Each group might use a different matching mechanism. 系统重新开放，并显示根据一定的配对机制所搭配好的所有的“学生-老师”配对。（在试点阶段，不必担心如何实施配对机制，因为配对设计组将运用matlab来进行这项工作，您只需要为配对组提供每个学生和老师的排位，并且告诉我们如何将搭配好的配对输入系统）。请注意，每个学生和老师都将会有一个群组ID。每个群组可能会使用不同的配对机制。

8. We will have several treatments. Some students and teachers will be matched with one mechanism not the other. We know some mechanisms have superior theoretical properties but we need to test which one is the most effective. 根据不同情况，我们将会有不同的处理方法。有些学生和老师的匹配需要使用特定的适合他们情况的机制。我们也知道，某些机制有超强的理论性，但我们需要检验哪一个对我们是最有效的。

Please write to

if you are interested.

如果您有兴趣，请写信至benjamin.chiao@gmail.com。