Paleonet: False PaleoNet Unsubscription E-mails
n.macleod at nhm.ac.uk
Thu Jan 5 07:49:27 GMT 2017
OK. Now that things have calmed down a bit it’s time to inform everyone on the list about what happened yesterday, how and why. First, what.
Yesterday around 18:30 GMT I received a message from the PaleoNet listserver informing me that I had been unsubscribed from the PaleoNet list. This came as a bit of a surprise insofar as I didn’t recall unsubscribing myself from my own list. But that being said, it’s not unusual for me to get the occasional unusual message from the list software as the list owner. I immediately checked the list and saw that, in fact, there had been no unsubscriptions of anyone yesterday. Since there was, apparently, no problem with the PaleoNet list I decided to wait and see whether this was a one-off internal glitch that came to me or whether it had been sent to any others. I didn’t have to wait long.
Shortly thereafter I began receiving messages from other PaleoNet subscribers alerting me to the fact that they had been sent an e-mail in my name informing them that they had been unsubscribed from PaleoNet and indicating to me they did not wish to be unsubscribed. Some subscribers thought they had done something wrong. Others thought it must be some glitch in the software. And others assumed they had been subject to some list-cleaning programme/routine. Once these started coming in I contacted Christian Jackson of NHM IT Services and asked him to look into the situation on the technical side as the NHM currently hosts the PaleoNet e-mail list on its servers.
As you might expect there is some background to this unanticipated event and, before I continue with the story I need to fill you in on that. On 20 September 2016 I was informed by NHM IT Services that a decision had been taken by the Natural History Museum to discontinue its hosting arrangement for the PaleoNet e-mail list and web pages. No specific reason for this discontinuance was given other than the following statement.
“Recently [NHM] IT infrastructure has been standardised and rationalised to provide more robust, sustainable and secure services to meet the objectives of the Museum. Specifically a major project has been carried out to re-platform the Content Management System (CMS) and re-launch the NHM website, closing many sites and moving others into new platforms (e.g. the Data Portal). As such the Technology Solutions (TS) department need to manage current and future web hosting services in line with the new systems and services in place to best optimise use of resources, and to do this TS will be establishing a clearer hosting policy.
The Paleonet site has been identified as one that the NHM is not able to provide a hosting and support service for going forward, and will need to be supported by alternative hosting arrangements.”
Ivan Teage, Digital Development Manager
The NHM has provided hosting services to PaleoNet since 2006. Since then PaleoNet has made very minimal demands on the NHMs IT infrastructure. There had been no indication to me that there was any lack of compatibility between PaleoNet IT requirements and new NHM IT infrastructure or between PaleoNet operations and the mission of the NHM. There was also no consultation between myself and the NHM over its hosting arrangement for PaleoNet prior to my receipt of Teage’s notification informing me of the museum’s decision.
I immediately sought advice from colleagues regarding migration of the list to a new host. Following discussions with Alan Spencer I decided to write a hosting proposal to the Council of the Palaeontological Association (PalAss) to migrate PaleoNet to their servers. That proposal was submitted for consideration at the Council’s October meeting where it was approved in principle. Between October and December the Palaeontological Association drafted a formal contract for the hosting of PaleoNet. I received a copy of this contract on 7 December and, after a bit of negotiation over wording, that agreement was signed on 9 December. I informed the NHM that an alternative hosting agreement for PaleoNet had been approved on 12 December.
The migration of an e-mail list is fairly straightforward but, inevitably, is a bit tricky as each list is interfaced within its host IT infrastructure in a slightly different manner. Also, the automated nature of listserver software is such that there’s no need for system administrators to become involved with listserver operations until/unless something goes wrong. Since this is infrequent in PaleoNet’s case it usually takes a bit of work to recall exactly how the list operates within the host IT server system and where the problem is. The migration from the NHM servers to the PalAss servers should be easier than the previous migration from the UC Berkeley servers to the NHM servers because the same listserver software that hosts PaleoNet now (Mailman) will be used after the migration. However, small glitches can, and often do, occur in the interval leading up to the migration (as system managers become familiar with the operation of the list-hosting software) and for an interval thereafter (as the listserver operations ‘bed in” with other operations going on within the new hosting server system). What happened yesterday was one of these glitches.
My plan had been to meet with Christian and Alan prior to the migration of PaleoNet to agree a plan and then to write to the list informing all of you of the migration, the reasons for it, and the migration plan. Part of that communication was going to be a caution that glitches might occur during the migration interval, along with a reassurance that full operations would be restored for everyone on the list following migration. Christian, Alan and I had a Skype meting planned for last week, but that, unfortunately, didn’t take place owing to technical issues on the NHM end. The plan for agreeing a plan was to pick up that conversation next week. However, it now seems as though an inadvertent glitch occurred on the PalAss server side as Alan was adjusting some of the Mailman software settings. The unusual source address that was part of the false unsubscription e-mails (paleonet-bounces at paleonet.org) was there because this message came from the PalAss - not the NHM - server. As near as Alan can tell something like 700 false unsubscription notifications had been sent out to PaleoNet subscribers before he realized what the problem was and killed the routine. That’s less than half the list, but still a large number. On behalf of PaleoNet I apologise for the surprise, dismay and confusion suffered by those who received false unsubscription messages and the inevitable confusion the messages about the effects of the glitch must have caused all subscribers. I wanted this migration to be transparent and seamless, but best laid plan often do have a tendency to become more complicated than you hope they will be.
So, there you have it. PaleoNet is leaving the NHM after 10 years of successful operation and moving to the Palaeontological Association servers. I’d like to thank the NHM for the service and support it has provided to PaleoNet - and through PaleoNet to the palaeontological community worldwide - for the past 10 years, and to thank the Palaeontological Association for agreeing to host and support PaleoNet into the future. Most of all, I’d like to thank you for your support of the list. PaleoNet is (literally) nothing without its subscribers. After being made aware of the NHMs decision to discontinue its support for the list I had been discouraged about PaleoNet’s future. Now, I am confident about it.
Since it’s inception in 1995 PaleoNet has become part of the core infrastructure of serious palaeontology, which has been very gratifying to see. Migration to the Palaeontological Association servers will provide an opportunity to further develop the list and the web site, which is something I’ve been wanting to do for some time now. There will, no doubt, still be the occasional glitch in PaleoNet operations in the future, just as their has been in the past. But the important thing - that PaleoNet provides and quick, easy, stable, informative, and fun forum for the discussion of palaeontological topics and dissemination of information about palaeontology - won’t change; at least as long as I’m running the list.
Thank you for being a PaleoNet subscriber.
With kind regards.
Professor Norman MacLeod
The Natural History Museum, Cromwell Road, London, SW7 5BD
(0)207 942-5204 (Office Landline)
(0)785 017-1787 (Mobile)
Department of Earth Sciences, University College
London, Gower Street, London WC1E 6BT, UK
Nanjing Institute of Geology & Palaeontology,
Chinese Academy of Sciences, 39 Beijing, Donglu, Nanjing, China
More information about the Paleonet