[by DDF (5th year physics grad student) and NKR (senior
undergrad)]
Begining... General Philosophy and Ideas Case Study: University of São Paulo Case Study: Brown University Summary BDH Letter
RMS (Free Software, Free Society) and ESR (of "The Cathedral and the Bazaar" fame)
{WebCT} vs {zope, squishdot, plone}
{MS Exchange} vs {qmail, postfix}
Linux Journal (May'04): Ludovic Marcotte took on the task of creating a new email infrastructure for a university with 35,000 users and 125,000 incoming messages per day. And up to 80% of the incoming email traffic is SPAM. OpenLDAP, Cyrus and Postifix came to the rescue.
+IRC (chats) +Newsgroups (usenet or Usenet Info Center Launch Pad)
What does this [all] imply/mean to our culture?
The following letter was sent to the Brown Daily Herald but, as of today (12Apr04) has not yet been published. Even though the BDH should publicize and be concerned not only with Campus issues but with bigger issues involving the whole Brown University and/or its Community as well, apparently it is quite hard of a task to make them publish something of value and content. On the 08Apr04 I have, personally, gone to the BDH office and talked to two of its editors. They have said that they are looking into the matter.
Dear BDH Editor,
My name is Daniel Doro Ferrante and I am the Technology Officer before the GSC (Graduate Student Council; [1]). I am also a 5th year graduate student in the Department of Physics.
What follows is of some length, a fact made necessary by the complexity and severity of the matter at hand.
I often hear that the present goal for Brown's IT is to become one of the World's leaders in technology as it once was back in the 80's. Personally, I am fully committed to helping this become a reality: "I want to do my best to help Brown make its dreams concrete!" As outlined in Brown's recent "Plan for Academic Enrichment", [13,14], I also think that the enhancement of undergraduate education, excellence in graduate education, innovative faculty research, multidisciplinary activities, enhancement of diversity, shared sense of community, diversification of the University's sources of revenues, collaboration with local community and the enhancement of our infrastructure and administrative support are key to Brown's [academic] success as a leading institution. Further, I also believe that Brown's strengths are unique and that we should build on them.
However, the IT decisions that have been made so far are clearly pointing in the opposite direction. Unfortunately, a number of recent resolutions by CIS and CAB (Computing Advisory Board, [15]) not only work counter to this vision, but also threaten several of the core educational and research missions of this University. This detrimental trend comes in the form of both specific, concrete policy decisions, as well as a more fundamental shift in philosophy behind the way decisions are being made.
Furthermore, these decisions have wasted Brown's money, allocating funds for solutions that are not suitable for this University, diverting them from more appropriate uses.
In addition to that, both CIS and CAB should be serving the community and not pushing it into some arbitrary direction. The University, its body and its community should be free to choose the better way for research, learning and teaching, goals whose achievement require support from CIS and CAB. However, as things currently stand, they are being prejudicial and detrimental to research, learning and teaching, therefore not doing the job that is expected (and required) from them.
Firstly, Brown was in need of an upgrade to its email service and, on top of that, it was decided that calendaring was an issue that should be implemented in the same solution. It is worth pointing out that the calendar feature is of little (if no) import to undergraduates, graduates and faculty. In spite of the fact that, right now, there are three calendars at Brown: calendar.brown.edu; MSExchange and WebCT.
As of 2003 it had been 8+ years that email services were handled by UNIX servers.
When this "upgrading process" began the idea was to keep this service in UNIX servers (as you can see in the attached documents) however, and I quote: "[CIS] is being strongly encouraged [by the new CIO] to seriously evaluate Microsoft Exchange as a server platform for these [email and calendaring] services."
The person in charge of this "changeover" was informed of the possible problems that an Exchange solution involved, among them: (1) vulnerability to viruses if local machines are not all kept fully patched (on top of its propagation due to Outlook holes, major downtimes and stripping of all attachments); (2) high cost (hardware and software wise, not to mention training and constant update/upgrade); (3) low efficiency; (4) [lacking of] anti-spam software; and (5) not clear that everyone will be willing to be in only one Exchange. In addition to these, the Exchange solution is not "standards [RFCs] compliant" (meaning the software does not correctly implement those policies whose existance are responsible for the Internet as we know it) which causes a myriad problems of different problems, including that of my students emails being caught by SPAM filters because of this non-compliance. Furthermore, Microsoft apparently has no answer for the robustness of a large-scale use of IMAP over Exchange (as of summer 03, as can be seen from the attached files) and Outlook used to have broken support for IMAP over SSL. Even peer institutions which are "MS-centric" seem to be reluctant on putting Exchange into production (due mostly to it's poor record respecting security, reliability and scalability).
This same person was also told of other possible solutions, like Sun based ones, GNU/Linux and other free and open source projects. In fact, Washington University invited Brown to participate in the development of their open calendar project [16] (again this is in the documents attached). Also, this person was told of peer places with a very diverse mix of solutions that basically included E2K/Outlook exclusively for mail and calendaring and centralized IMAP (on UNIX) that serviced the mail.
It is clear to me that Brown should be making its ideas concrete (expressed in its mission statement and on its plan for academic enrichment) by means of actions (and vice-versa), however, these current decisions are not backing up the University's plan for "multidisciplinary" initiatives, excellence through diversity, building of a shared sense of community nor enhancement of the quality of our infrastructure and administrative support.
On the one hand, while I do understand the different types of problems faced by CIS and CAB, I cannot agree with their decisions, after all, the University has chosen to be part of the "bleeding edge", i.e., we chose the "hard way". And, sure enough, the decisions we will be facing - going down that road - are not going to be easy. They are supposed to be complicated, intricate, difficult and even hard. That is the choice made by the University and by us, members of the Brown Community. And not only did we accept that, we are willing to collaborate, to participate, to help in any possible way. But, in order to do that, we cannot be handicapped by the institutions whose jobs are to facilitate our interactions, our ideas, our dreams.
At this point, I am basically talking about a philosophical change. It is quite clear that we cannot expect to have "support" for our technologies, because they will be innovative, forward-looking, and diverse. The whole Brown Community has to be part of the "support", we will be involved, we will make the dreams laid down in the University's Mission and in its Plan for Academic Enrichment come true, we will be the instruments that will make it happen. But, again, we cannot be limited by the very institutions who should be working with us, on the same team, in order to make this an ever bolder achievement. (Furthermore, these institutions should not be looking for means to "blame someone else" when dealing with support and like questions. They should engage themselves and do the work that is expected from them. This attitude of "finger pointing" is extremely counter productive and Brown should be constituted of people that can see further than that.)
The same behavior encountered above can be found with Brown's "e-learning" solution, WebCT, [2]. This solution is expensive and not flexible and dynamical enough to fit Brown's needs. Basically, WebCT offers one a static platform with pre-made choices. Moreover, the interface is cumbersome and does not permit much flexibility for the interaction between teacher and student.
Other options are [far] cheaper, more dynamical and could have been tailored to Brown's specific needs in a much more rich manner. For example, a combination of Zope, [3], Squishdot, [4], and Plone, [5], creates an extremely diverse environment that is capable of providing and handling the edge that Brown needs in order to be a leader among its peers. Furthermore, providing HTML-based forums is an outdated, slow [and "heavy"], insecure, unreliable and of poor scalability solution, even though it is one still widely used. Although they are aging at this point, Newsgroups (Usenet) and IRC (Internet Relay Chat) are lightweight, dynamical and of proved scalability tools that can enrich our environment when used properly. The pop-culture massacre that leads to HTML-based answers is not usually justified.
Further, if we are seeking speed of content delivery, we should be using flexible solutions. Students come to Brown because they are looking for a great education, for a diverse environment, for a free atmosphere and, fortunately or not, this is not always done via some new technological toy! The best education can also be done on the blackboard; a diverse and multi/interdisciplinary environment is built on the interactions among the community members and a free atmosphere is nourished by the combination of all this. Brown needs great teachers: they make a difference, technology only helps. In fact, all that this new "JIT" [Just In Time] education "concept" does for a teacher is to enable him/her to disseminate his/her "teaching style" in a different envelope. Without a solidly good teacher, that also knows how to use the technology, it is just eye-candy. JIT e-learning is not going to make the average teacher get any better, at most it can make him/her look less mediocre and fool students for a bit longer.
As a school, a "universal learning place", Brown needs to have faculty that are committed not only to research but to education as well. These are people that will be part of the Brown Community, that will work to make it better and better. Those people will have the qualities described and chosen not only by Brown's Mission but by its new Plan for Academic Enrichment too. We cannot expect that new technologies will make our environment more diverse, interdisciplinary, more dynamical, more enriching, more nurturing of new ideas! Our Community has this job: Our undergraduate and graduate students, our Faculty, our Staff and everybody else that is part of this "place" called Brown University. That is how a strong, bold and resilient leading institution is born.
After all, those will be the people involved in research at Brown. They will be leading us towards the future, one that we have already chosen as being innovative, new and pleasant, but one that will bring us difficulties and hardships never faced before, decisions that will not always be clear cut and easy to make, paths that will not be clear at times. However, this is the price that we, as Brown Community Members, are paying, and we are paying it consciously, always bearing in our minds that we chose it for a reason. And this reason is that we want to perform as best as we can.
Universities are [historically] supposed to be an open environment, where the freedom of ideas and thoughts is paramount. Our IT choices and decisions should reflect that. We should be free and open with our IT choices and decision making procedures. Research has always pushed IT resources to its absolute limits (software engineering, writing and designing and systems creation and management) and, thus, our environment here at Brown should reflect that. New policies (such as [8]) clearly do not help in achieving the goals set forth for Brown. On the contrary: they interfere and possibly prevent the very best research from happening.
It is clear that Brown is moving towards a more centralized structure. This type of organizational scheme tends to be more rigid than its decentralized counterpart. This rigidity, in turn, will deprive Brown of the flexibility and dynamism that we are all seeking and, even more, that a research environment needs in order to keep itself flourishing. Our choices should be left open, in fact, and the University should look for solutions that would encompass this freedom, for it is the only way to keep the meaning of the word "university" alive.
I believe that Brown should be getting students (and the Brown Community in general) involved into this goal, just like Berkeley and MIT have done in the past (and in one case it ended up giving birth to BSD, [9], OSX, [10] and the SETI@home project, [11], and, in the other case, it gave birth to the GNU, [12], and FSF, [12], Foundations)! We should have home-grown software/solutions, we should jump into new projects that have not been thought of before. Brown could make better use of newsgroups and of IRC (as a means of chatting) in order to better integrate its community: Faculty, undergrads, grads, staff, etc... We could be engaging on much bigger projects, collaborating with other places and peer institutions! There are no limits to this, just our imagination, enthusiasm and freedom.
Even though I do understand the need for stability, reliability and resilience of a central system, there are ways to accomplish this without losing flexibility, dynamism nor creating "chaos" and disrupting the University's services.
As the situation stands right now, I do not believe Brown's dreams to be possible. On the contrary, given the current status we will have to make quite intricate changes if we want to put Brown back in the bleeding edge of IT. I believe that the example put forth by our peers here in the USA and in other countries (from Switzerland to Brazil) is paramount to our guidance and also to tell us about our current standing.
Thank you very much for giving me this opportunity to speak my mind on this subject and also for bearing with my strong feelings.
Yours sincerely,
Daniel Doro Ferrante.
Refs.:
[1] GSC
[2] WebCT
[3] Zope
[4] Squishdot
[5] Plone
[6] PH050
[7] PH232
[8] Network
Draft Policy
[9] BSD
[10] Mac OSX
[11] Seti@Home
[12] GNU
[13] Plan
for Academic Enrichment (News Bureau)
[14] Plan
for Academic Enrichment (Official document)
[15] CAB
[16] UCal