Professional Staff Congress | 61 Broadway, 15th Floor, NYC 10006 | 212-354-1252 |212-PSC-CUNY | firstname.lastname@example.org | AFT Local #2334
CUNYfirst, Users Last
Professor of Computer & Information Science
[What’s your experience with CUNYfirst? We’d like to hear about it. Click "add new comment" at the bottom of this article to tell your story.]
Every once in a while I get a question, either privately or in a department meeting, regarding CUNYfirst. Here is what I know of CUNYfirst, based on a few years of working with the project as a “training liaison” (which is a fancy term for room-scheduler):
1. The idea of CUNYfirst is a good one: to have a unified, integrated “enterprise”-scale system that encompasses all university/campus business processes. Such a system could, in principle at least, have saved a lot of expenditure on maintaining dozens of disparate, redundant, barely cooperating third-party systems. Such a system could have offered information access that would have benefited the administration, the staff, the faculty and the students.
2. However, CUNY Central’s motives in pursuing CUNYfirst were dominated by an agenda that is quite apart from such benefits. Rather, CUNY Central sought absolute control over all college activity, including curriculum. For example, whoever controls the catalog, the bulletin, the transcripts and the apparatus in general effectively controls curriculum. CUNYfirst will be part of the arsenal by which CUNY Central shoves Pathways down our throat. CUNY Central also sought the knowledge of, and therefore access to, any discretionary funds that the colleges may have.
3. The negotiations that were the run-up to the purchase of CUNYfirst were a travesty. The project required an expenditure of up to a billion dollars to do it right. CUNY Central offered far less. All but one of the bidders dropped out as a result: the project could not be done properly with what CUNY offered. Oracle-PeopleSoft did not drop out. However, they warned CUNY that for that level of funding, they could not, would not customize – they would only configure.
CUNY Central was so eager to have a centralized MIS [management information system] tool to use for its own centralizing, corporatizing agenda, that it totally ignored the implications of the Oracle “configure-only” limitation: business processes would have to be made to fit Oracle, not vice-versa. Capabilities that we now have will vanish. The staff, the faculty, the students would just have to “adjust” (the non-technical term being “suck it up”).
4. No large computer system is perfect, and there are trade-offs that come with each decision. CUNY could have addressed the challenge of having limited funds by narrowing the scope of the project somewhat. That might have made it possible to consider the widely used and highly regarded Banner system currently provided by Ellucian. Banner was developed for university environments from the start, so costly customization would have been less necessary. However, CUNY limited its own options because of its obsession with an all-encompassing, centralized system. In any case, Banner is not the only academic-oriented ERP [enterprise resource planning] system; there are others.
5. People knowledgeable about CUNYfirst say that about $600 million dollars has been spent so far – on a system that makes things worse. The actual cost far exceeds the $600 million dollars that goes to Oracle. Because processes are now much more inefficient, more people have to be hired to do tasks that were formerly automated or more burdens are placed on HEOs, clericals and even administrators on the executive plan.
Unknown and unseen to most faculty has been the toll that this takes on HEOs and to some extent clerical workers – the people who actually make the university run. (No, professor colleagues, you are important but you do not make the university run – though that is another discussion.)
HEOs have been forced to put in all sorts of extra hours – and too often have not received compensation. Some of this work is transitional, but some of it is systemic.
6. CUNYfirst does work. It just works badly.
- The interface technology is laughable: it has the appearance of an early-’90s update of 3270 bi-synch technology. Web 2.0? Ha. Not even web 1.0.
- The security model is totally inappropriate for CUNY: we will have work-study students performing tasks that require vast permissions, thus allowing them to access data of other students. When I log in to my faculty account, I am able to get names, birth dates and the last 4 digits of Social Security numbers of hundreds of students whom I have nothing to do with, who are not in my classes, my discipline or my college.
- HR has had to struggle with the “problem” of an individual being a grad student enrolled in one campus, serving as an instructor at another campus and having a part-time office job in a third. GM and Apple don’t work that way. But CUNY does.
- Because CUNY wouldn’t pay for customization, we had to renumber our courses. This is just one of many changes less visible to faculty that CUNYfirst has forced.
7. Consider the simple task of a professor logging in to see the current roster of a course. It takes no fewer than five mouse clicks after logging in (never mind that the login process is disrupted by website security certificate problems – temporary, one hopes, but emblematic of IT incompetence). In the old Brooklyn College portal, only two clicks were required. This seems like a small thing, and for faculty there really are no consequences. But for HEOs and clericals who spend all day clicking and entering, increasing the number of user actions by a factor of 250% will have an impact.
And it gets worse. There are clicks and there are clicks. Brooklyn College clicks (from my off-campus, off-CUNY computer) typically take under two seconds. CUNYfirst clicks require four or more seconds. The total time it takes me to log in to Brooklyn College, and get my roster displayed is 20 seconds. For CUNYfirst it is 80 seconds. Again: I’m a professor and who the hell cares whether it takes me another 60 seconds to get a roster? But HEOs and clericals use this all day long: a system that is at least four times slower than it ought to be.
And this is not all. The user interface is an affront to common sense and guarantees a need for extensive training for administrative users. For example: I log in. The system “knows” that I am a professor. Why then am I confronted with a dozen links, half of which have no relevance to my role? The links themselves are confusing. I’m searching for my class roster. Why would I expect that functionality to be found in “Self Service” rather than “Records and Enrollment” or “Campus Community” or “Reporting Tools” or “People Tools” or “CUNY”? If I click on “CUNY” I see it has a subsection “Campus Solutions” – wouldn’t that be a place to find rosters? After a while, you do find what you are looking for. (Hint: when you open up “Faculty Center” don’t be so foolish as to excitedly click “Class Roster” – you must first click “My Schedule” and then find the unlabeled graphical icon that looks like three upper torsos, and click that.) Thus, CUNYfirst is a fabulous online version of the Where’s Waldo? books.
8. We at Brooklyn College (and other “Wave 3” campuses) will adjust. I know people in other schools (in earlier “waves”) who have. We will suffer more than they have because Brooklyn College has had the best add-on systems (for scheduling, grade reporting, etc.) of the university. Many of these will now go away.
9. HEOs say the system is frequently down, and for prolonged periods of time, which requires the double-work of writing information down on paper, again and again, to be entered later in the computer. These problems will hopefully be resolved over time – but right now we can only hope.
10. I witnessed and participated in some of the early end-user testing. That’s the phase of the testing process where all the basic elements of a system are supposed to be working, and the goal is to identify potential anomalies resulting from complex sequences of real user activity, activity determined by the customer. Instead, we followed a test script provided by the vendor. We never got to “anomalies resulting from complex sequences”: the system failed on the most trivial actions. As it failed multiple times, an engineer from Oracle would run to the next room to adjust something and then we testers would retry. This was a totally inappropriate methodology and in complete violation of long-standing software development practice. I’ve heard, though, that they’ve improved this process somewhat.
11. As CUNYfirst extends its reach across the University, the thing to keep in mind is that no matter how bad CUNYfirst is for YOU, CUNYfirst is a success for CUNYCentral.
What’s your experience with CUNYfirst?
We’d like to hear about your experience with this new computer system. Click "add new comment" to tell your story. (The only criteria for posting is that comments be on topic and civil in language.)