Exec meeting yesterday morning- one of the main things we discussed was Business Continuity and incident issues. The possibility of more student unrest and industrial action by staff, no matter what your political alignment might be, give us management issues which we need to think about in advance and have plans for. And as well as possible incidents, there's always real ones to deal with! Last week a planned change to our DNS was followed by something falling over - live solaris zones? a solaris cluster? - not sure of the terminology, but it was Bad. As usual, I was away (often am when Bad Things Happen) so they were able to deal with it without me interfering. However, yesterday when the same machines decided to have another unscheduled rest, I was here, so I could interfere. That usually consists of me wandering into the infrastructure team's office and asking silly questions like, Do you know what's happened?, and Can you fix it? Sometimes I look at screens knowledgeably and listen, although you can hear some strange things, like "Pussygalore's not responding". I'm hoping it's the name of a server...
So, another incident review this afternoon where we'll be doing the usual what went wrong, how can we deal with it, what lessons have we learned etc.
Later in the afternoon we met with an academic department to discuss some service and operational issues they'd raised with us. I like to get out into departments, but it doesn't happen as often I'd like - things are always easier to discuss and resolve face to face.
This morning spent mainly catching up, reading emails and notes of meetings, and some new blog posts. You might remember me discussing an institutional review of twitter in Russell Group Universities which Brian Kelly had posted about on his blog. Now the study has been repeated, but looking at the 1994 Group of Universities. Wide range of results, with some Universities making more use of it than others. Not had chance to see if there's any significant differences between the two groups yet, but there's some useful data there for people managing institutional twitter accounts.
4 comments:
I think pronounceable server names are much easier to reference during an incident rather than talking about problems with eg uiwwwsci01, or did you say uiwww01?
Interesting to see @sheffielduni accounting for almost half of the "following"/follow backs for the Russell Group institutions.
Is it possible to analyse the number of followers who are internal to a given institution?
i.e. to what extent does its use reach the outside world?
It would also be useful to know how institutions make use of Twitter:
PR/External relations?
To what extent do institutions follow/connect/interact via Twitter? Of course tricky in that this would largely be via personal accounts?
Realtime service announcements?
Is it possible to get metrics for "click throughs" from Twitter announcements. e.g. If a Twitter post refers to a web page can the hits due to seeing it on Twitter be counted? i.e. to guage the effectiveness of Twitter v.s. other channels.
How do institutions publicise the availability of a Twitter channel and what's the relative take up?
Do institutions integrate Twitter with their IT systems?
Looking at the raw numbers, it doesn't seem to say much beyond usage is low, and take up is extremely poor given the numbers of staff, students and external people who interact with insitutions compared to the number following the institutions on Twitter.
"So, **another** incident review this afternoon where we'll be doing the **usual** what went wrong, ..."
Not entirely sure that conveys a great impression!
Agreed, it doesn't, but I'm being honest. We have had a number of incidents over the last year - too many and I don't like it, but it is a fact. Of course, some of the incidents we've reviewed have been outside of our control - the snow, and the student occupation for instance.
Post a Comment