Workflow for adding new content to LOCKSS

Every few weeks LOCKSS Alliance members receive a message informing them of the release of the latest batch of Archival Units (mostly journal volumes). At De Montfort University we do not respond immediately, but prefer to wait for the LOCKSS website’s new page to be updated. From the website we download the latest batch details as a spreadsheet. For example, in the first week of March we are working on this spreadsheet with 309 New Volumes. It is a list that usually contains journals to which we are anxious to ensure continuing access, and others to which we either have no preservation rights or no such anxiety.

There are plans for adding new features to LOCKSS in the Summer of 2013 to make activating new content easier. For now, our workflow is as follows:

To sort the LOCKSS list we look to our Electronic Journals A-Z and OpenURL Resolver service SFX, from ExLibris, for a list of all our live holdings. We export all this data into Excel and sort it by ISSN number. Large sections of this list can be ‘pruned’ to remove packages where there are no preservation rights, for example: Proquest, EBSCO, etc.

With the SFX list down to a manageable size, we can use the spreadsheet’s VLOOKUP function to merge data from the two files. We create a new column in the LOCKSS list and match in the threshold data from SFX. Where there is no imported threshold we can delete rows in the LOCKSS list as we would have no rights or interest in these. If the list is short enough, it is worth looking over these matches and removing lines where the LOCKSS AU is outside the subscription threshold range.

Now you could use this list to activate AUs in LOCKSS, but at DMU we go for a further stage and involve subject librarians in the selection stage. The amended list is passed on to them for checking. Only when it has been reviewed and returned do we begin activating the new AUs in LOCKSS.

Once a month, at least, we update SFX, using its ‘dataloader’ with the collected LOCKSS journals and thresholds. It is only at this point that all the effort put into preserving electronic journal content is visible to our users.

Posted in Digital Preservation, OpenURL | Comments Off on Workflow for adding new content to LOCKSS

Assessing the coverage by LOCKSS of Free Electronic Journals

LOCKSS, as a electronic journal digital preservation system, covers subscribed and freely available journals. I have been looking at the balance of activated titles at De Montfort University and found about 43% to be subscribed titles and 27% to be journals listed in the Directory of Open Access Journals.
A further category of journals would be those available without charge, but not included in the DOAJ. These would be included in the SFX Free ejournals target. However there might be an overlap between this target and the DOAJ.

Num. of Titles Unique Title Overlap % Unique Titles % Title Overlap
Set 1 676 557 119 82.4 17.6
LOCKSS 676 557 119 82.4 17.6
Set 2 20479 20360 119 99.4 0.6
Free E- Journals 20479 20360 119 99.4 0.6

The overlap between LOCKSS and DOAJ/Free Ejournals combined in 276 titles or 40.8%.

Posted in Digital Preservation | Comments Off on Assessing the coverage by LOCKSS of Free Electronic Journals

Assessing the coverage by LOCKSS of subscription journals

A previous post looked at the count of Open Access journals being preserved in the De  Montfort University LOCKSS service. The figure came out as 27% of LOCKSS titles being represented in the Directory of Open Access Journals.

It may be worth looking at the figures for subscribed journals as well. To do this I used the same SFX Collection Tool to look at the overlap between activated LOCKSS titles and packages of journals where a subscription is required.

Num. of Titles Unique Title Overlap % Unique Titles % Title Overlap
Set 1 676 384 292 56.8 43.2
LOCKSS 676 384 292 56.8 43.2
Set 2 763 471 292 61.7 38.3
BMJ Journals Online Archive 6 2 4 33.3 66.7
Edinburgh University Press 3 0 3 0.0 100.0
Emerald Current 62 43 19 69.4 30.6
Emerald Management Xtra 111 162 77 85 47.5 52.5
Highwire Press American Association for Cancer Research 3 1 2 33.3 66.7
Highwire Press American Association of Immunologists 1 1 0 100.0 0.0
Highwire Press American Speech Language Hearing Association 6 6 0 100.0 0.0
Highwire Press Journals 28 13 15 46.4 53.6
Highwire Press University of Wisconsin Press 1 1 0 100.0 0.0
MetaPress Journals 17 13 4 76.5 23.5
Miscellaneous Ejournals 106 105 1 99.1 0.9
Oxford University Press Journals 51 11 40 21.6 78.4
Portland Press 2 1 1 50.0 50.0
SAGE Communication Studies Full Text Collection 19 0 19 0.0 100.0
SAGE Complete 114 29 85 25.4 74.6
Springer Complete Collection 94 93 1 98.9 1.1
Taylor & Francis Journals Complete 142 106 36 74.6 25.4

In Set 1 there are the 676 titles archived in LOCKSS and a reported 43.2% match between these and the titles collected from a variety of publishers in set 2. The second set does not list all the subscriptions taken at DMU, so one should be wary of concluding that 38.3% of such material is archived in LOCKSS. The main interest in this report is in the lines for each publisher and their coverage in LOCKSS.

  • Title coverage is complete for Edinburgh University Press and the Sage Communication Studies package.
  • Title coverage is at 50% or over for BMJ, Emerald Management Xtra, American Association for Cancer Research, Highwire Press, OUP, Portland Press and Sage Complete.
  • Title coverage is low for Miscellaneous Ejournals and Springer.

It should be recognized that just because a title appears in LOCKSS, it does not mean that every relevant article has been archived. There is further work to be done on examining the range of extent of LOCKSS Archival Units for each selected title.

There are also more titles in LOCKSS not included in this or the DOAJ report, so these will still need to be identified before a full picture of LOCKSS coverage at DMU can be determined.

 

Posted in Digital Preservation | Comments Off on Assessing the coverage by LOCKSS of subscription journals

Assessing the coverage by LOCKSS of open access journals

At DMU we have an electronic journal digital preservation system using LOCKSS and an OpenURL Resolver using SFX that makes information available to people on which journal they can get access to and from where. We are the first university in the UK, at least, to get these two services talking to each other.

This has an advantage for staff and students at the university who can see and use the preserved journal articles when they are needed. I also gives us the chance to start assessing how running the LOCKSS service is helping to guarantee continuing access to the journals we value in the university.

For example, we can compare the Open Access journals recognised in the Directory of Open Access journals (DOAJ) with our local holdings in LOCKSS. SFX has a Collection Tool that enables this.

Of the 676 journal titles represented in LOCKSS at DMU 188 or 27% are journals in the DOAJ. More than a quarter of the journals being collected and preserved in LOCKSS are open access journals.

On the other side, of the 8524 journals in the DOAJ, that same 188 represents 2.2% of the journals. That covers the actual preservation work we are doing with this set of journals. The potential is likely to be higher, but we do only select the journals of interest to the mission of the university. If we were to attempt the preservation of DOAJ journals without discrimination we would find that 559 or 6.4% of journals could be covered.

This does look like at interesting area for further investigation. I can see a further blog post is called for taking this farther.

Posted in Digital Preservation | 1 Comment

Nested lists in XHTML

There are lots of examples across the Internet explaining how to use CSS to number items on a list. It is certainly worth trying this technique, rather than hard coding the numbers in, or using the now deprecated<start=”15″>, if only to make it easier when you need to remove items later on.

The numbering system I wanted to achieve did not seem to be used as an example anywhere else, so I am logging my solution here: a) in case it is useful to anyone else; b) so I don’t forget it myself.

What I wanted to achieve

0. First item

a. first sub-heading

b. second subheading

1. Second item

a. first sub-heading

b. second subheading

How I tagged the text

<ol class=”regulations”>
<li>Preamble/introduction
<ol>
<li>The Department of Library Services (hereafter referred to as the Library) provides users with materials, services and facilities for learning, teaching and research purposes. The regulations are designed to ensure that such provision is made as effectively and fairly as possible and that the integrity of the library service is maintained. The regulations cover the conditions for admission and use of materials and facilities, the prohibition of inappropriate activities and the penalties for infringement.</li>

<li>These regulations have the been approved by the Academic Board and have the support of the Governing Body of the University. They are subject to review from time to time.</li>
</li></ol>

How I set the CSS

/* Library Regulations formatting */
.regulations
{
counter-reset: heading -1;
}
.regulations ol
{
counter-reset: item ;
margin: 1em;
padding: .3em;
}
.regulations li
{
display: block;
margin: .5em 0;
padding-left: 2em;
}
.regulations li::before
{
margin-left: -2em;
margin-right: .5em;
text-align: center;
counter-increment: heading;
content: counter(heading) “.”;
}
.regulations ol li::before
{
margin-left: -2em;
margin-right: .5em;
text-align: center;
counter-increment: item;
content: counter(item, lower-alpha) “.”;

How it works

I am using the counter technique, as explained on numerous sites like: Zibaldone and W3schools. The ‘counter-reset: heading -1;’ reset the counter for ‘heading’ to the number below zero, so that when the ‘regulations class is encountered the count will increment to zero.

I use: ‘counter-reset: item ;’ to reset the counter for ‘item’ to the default number which is already zero. There are some other formatting instructions in ‘.regulations ol’ and ‘.regulations li’, the most important of these is the ‘display: block;’.

In ‘.regulations li::before’ I have:

counter-increment: heading;
content: counter(heading) “.”;

These two lines add one to the Headings numbering sequence and display the result, followed by a ‘.’.

To get the sub-headings to display preceded by a count using lower case alpha letters (a,b,c, etc.) I use:

counter-increment: item;
content: counter(item, lower-alpha) “.”;

in

.regulations ol li::before

 Result

You can, for the moment, see the finished result on the Library Regulations page on the DMU Library website. However, plans are afoot to revise and re-publish these regulations.

Posted in XHTML | Comments Off on Nested lists in XHTML

Towards a SMART Goal for LOCKSS

I mentioned in another post that the number of times LOCKSS appeared as an option on the ‘Find It @ DMU’ menus generated by SFX, our OpenURL Resolver was an interesting figure to examine. I only have one month’s worth of data so can’t report on an upward or downward trend yet, but let’s not let that stop us. For the whole month of December 2012, LOCKSS showed up on 6.8% of the menus.

It would be helpful to have a SMART objective for monitoring the growth and usefulness of this service.

Suggested objective: LOCKSS should show up as a menu option for 7% of the menus generated by SFX.

To check whether or not this counts as a SMART goal, I am checking it against the criteria highlighted in the Wikipedia article on SMART Criteria.

Specific

A specific goal will usually answer the five “W” questions:

  • What? I want to increase the proportion of material included in the library’s activity designed to ensure continuing access in the long-term to  electronic journals.
  • Why? To mitigate the short and long term threats to electronic journal information, and to show that LOCKSS is a tool relevant to such an endeavour.
  • Who? Content Delivery team members in the library who activate Archival Units in LOCKSS, opening them to preservation; Subject librarians who select such material for preservation; staff and LOCKSS in Stanford and the UK LOCKSS Alliance who liaise with publishers and complete the technical stages of enabling preservation to take place; publishers of electronic journals who take on to share the responsibility for ensuring continuing access to electronic journals (a part of recorded knowledge) with libraries.
  • Where? At De Montfort University: specifically the interface between LOCKSS which does the preservation and the OpenURL Resolver SFX which enables access to the preserved material.
  • Which? Relevant material needs to be available for archiving in this way; selected by the library; collected by LOCKSS and access enabled in SFX. There is an element where the user is involved in selecting which menus are generated by the OpenURL Resolver, so archiving lots of journals in which the users have no interest would not advance progress towards this objective.

 Measurable

A measurable goal will usually answer questions such as:

  • How much? 7% of the menus generated as a result of user requests to the OpenURL Resolver, which list the ‘appropriate copies’ to which the user has access, should include the option of getting the material via LOCKSS.
  • How will I know when it is accomplished? Regular checking of the statistics produced within SFX.

Attainable

  • How can the goal be accomplished? By activating relevant material in LOCKSS for archiving; ensuring that it is successfully collected; ensuring that SFX is up-to-date with its record of material held in LOCKSS and made available to library users. There is also a requirement to continue identifying at-risk journals where preservation is inadequate and the material is of interest to members of the university.

Relevant

  • Does this seem worthwhile? It involves taking action now for the benefit of others in the future.
  • Is this the right time? The content available for archiving in LOCKSS is growing all the time. Membership of the UK LOCKSS Alliance is based on a subscription and it is important to have evidence of the value of this subscription when making decisions about whether to continue such membership.
  • Does this match our other efforts/needs? The library makes content available to users; enables this access through an authentication systems, measures and compares the usage for different services, checks that usage is within that permitted by the licenses we sign: LOCKSS archiving is a step towards guaranteeing continuing access and mitigating threats against this service.
  • Are you the right person? I am arguing that I am.

Time-bound

  • When? Good question, as I don’t know the rate LOCKSS content at DMU is growing or likely to grow. So let’s say within one year (end of 2013).
  • What can I do 6 months from now? Ensure that the service is valued by library managers to the point that they wish to continue providing (and budgeting for) the service.
  • What can I do 6 weeks from now? Ensure accuracy of holdings records for archived material, and that relevant material is being selected for archiving by librarians.
  • What can I do today? LOCKSS is already out of synch with SFX as new material has recently been collected. This has raised the number of journal in LOCKSS from 673 to 708 and other existing titles are likely to have wider date ranges successfully collected. This is already a positive step towards the stated goal – but only if SFX is updated to reflect the new holdings.

How will we get on in meeting this objective? Watch this space.

Update: January 2013

1531 appearances on SFX menus, out of 23092 = 6.63%. By the beginning of February 677 titles were activated on SFX.

Update: February 2013

1483 appearances out of 24021 = 6.17%. During March we will be activating thresholds for more 2012 volumes, so I would expect to see an increase once these are collected and activated, I hope. By the start of March 682 journal titles had been activated on SFX.

Posted in Digital Preservation, OpenURL | Comments Off on Towards a SMART Goal for LOCKSS

Prospects for expanding LOCKSS coverage of electronic journals

After just a few weeks of exposing the content collected in the university’s LOCKSS electronic journal digital preservation system it is interesting to look at the potential impact of this service. One way of doing this would be to check on the logs of SFX, the system we use to create our journals A-Z list and resolve openURL requests. SFX works by generating an individualised menu for each request it receives: it matches the user’s request with information, not just about the journals title, but also the annual coverage of our subscriptions.
SFX has a query that finds the ‘Top target services shown in the SFX menu’. Over the 1-16 December 2012 period LOCKSS came out as the 16th most frequently offered optional menu option. It is heartening to see LOCKSS jumping straight in to the top 20.
Another way of looking at this is to calculate the frequency of occurrence, see earlier post. On this measure LOCKSS comes out as being relevant to 6.7% of the requests received by SFX.
It is not clear whether this figure is high or low. But the same report givers some hints about the prospects for the figure to grow over the coming year.

SFX Services more popular than LOCKSS

There are full text services more popular than LOCKSS. However, most of these are either aggregator services like EBSCOhost or Proquest, who do not really have an archiving responsibility for their content. The only publisher on the list is Elsevier’s Freedom Collection. This publisher, according to the Keeper’s Registry, has digital preservation arrangements with other agencies.

SFX Services about as popular as LOCKSS

LOCKSS showed up on 835 of the menus generated in early December. There are a group of full text services between 100 and 1000 which I am calling ‘as popular’ as LOCKSS. They include direct publishers like Emerald and Wiley, and indirect groupings of publishers like the Directory of Open Access Journals (DOAJ) the free journals in Pubmed. There are prospects for growing the relevance of LOCKSS with this mid-range group of publishers. For example, at present only a few Taylor and Francis journals (and only a few years for each journal) are archived by LOCKSS, but this is likely to grow, even in the next 12 months. There are publishers producing journals that we care about either because we subscribe directly to them, or via open access gold publishing contribute to them.

The long tail of smaller publishers

There is a further group of SFX services appearing on 1-100 menus generated for the selected period. Some of these are for open access publishers that might already overlap with services mentioned earlier: DOAJ and Pubmed. Others are for publishers where we only have a few subscriptions, like MIT Press.

Expanding coverage

As new publishers make content available to LOCKSS the relevance of the material collected is likely to to grow for members of the university. Disk-space requirements are going to grow at a similar pace. It is only just becoming possible to visualise the part the LOCKSS may play in guaranteeing the security of access to the journals relied upon for teaching and research in the university. To really expand beyond the 10% mark we may have to look at supporting other digital preservation agencies, like CLOCKSS, to avoid the risk of having all our eggs in one basket.

Posted in Digital Preservation | Comments Off on Prospects for expanding LOCKSS coverage of electronic journals

LOCKSS coverage at De Montfort University

De Montfort University has been involved in the UK LOCKSS Alliance since its formation in 2007. Since then we have been building up the electronic content preserved there, but during 2012 it became possible to demonstrate the extent of the material collected. By the end of November 674 preserved journals had been activated in SFX, our OpenURL Resolver and journals A-Z list (most people know this as ‘Find It @ DMU’).

Over the roughly two-week period 1-16 December 2012, 12364 requests passed through SFX, each generating an individualised menu screen. The purpose of an OpenURL Resolver, like SFX, is to point people to ‘appropriate copies’ of the article they are looking for. The challenge is to match the request with the right source (where the user has the correct access rights) and the right date range (where their university has a subscription in place or other reason why access would be granted).

After probing the usage statistics it was found that 838 of the menus generated by SFX included LOCKSS as a target. This comes out as 6.77%.

838 / 12364 X 100 / 1 = 6.77

What conclusions can be drawn from this figure? LOCKSS preserved materials are clearly becoming visible to people using ‘Find It @ DMU’. Is that because librarians are preserving relevant material or users are asking for stuff that LOCKSS has preserved? It is also probably too early to compare this with other institutions running LOCKSS as not many have integrated LOCKSS with their OpenURL Resolvers – though I would be interested in hearing from anyone with similar figures.

It is possible to compare the popularity of LOCKSS with other full text services on SFX in the same period.

Position Target service Requests
14 EBSCOHOST_ACADEMIC_SEARCH_PREMIER getFullTxt 1903
15 EBSCOHOST_BUSINESS_SOURCE_COMPLETE getFullTxt 1564
16 NESLI2_ELSEVIER_SD_FREEDOM_COLLECTION getFullTxt 1304
17 PROQUEST_ABI_INFORM_GLOBAL_NEW_PLATFORM getFullTxt 1290
18 NCBI_PUBMED_FREE getAbstract 1034
19 EBSCOHOST_ELECTRONIC_JOURNALS_SERVICE getFullTxt 995
20 LOCKSS getFullTxt 838
21 EBSCOHOST_CINAHL_PLUS_WITH_FULL_TEXT getFullTxt 710
22 MISCELLANEOUS_FREE_EJOURNALS getFullTxt 607
23 WILEY_ONLINE_LIBRARY_JOURNALS getFullTxt 525
24 DOAJ_DIRECTORY_OPEN_ACCESS_JOURNALS_FREE getFullTxt 433

After just a few week with a full list of LOCKSS archived journals appearing on SFX, this service is in 20th position in the ‘top services shown in the menu’ chart. If you exclude the top four services, which have to appear in any menu, then LOCKSS is in 16th place.

Targets more popular than LOCKSS include mainly aggregator services like EBSCO and Proquest, with the exception of Elsevier. Targets less popular include the publishers represented in in LOCKSS, but this is not surprising, if you think about it.

DMU has continued to add content to LOCKSS where we think we would miss it if it were to become unavailable elsewhere for any reason. The effect is to add a kind of guarantee to the links made in reading lists and assignments.

Posted in Digital Preservation, Statistics | Comments Off on LOCKSS coverage at De Montfort University

Tuning Mendeley to find electronic journal articles

Mendeley, according to its website “is a free reference manager and academic social network that can help you organize your research, collaborate with others online, and discover the latest research”. If you are using this website to organise your references, you may be interested in tweaking to make it easier to find electronic journal articles. I am explaining here how members of De Montfort University can enhance their Mendeley accounts, but if you are with another institution, these notes should still be helpful.

Some artilces are shared within Mendeley, others cited with links pointing outside the site. There are often lots of versions of an article stored across the Internet, but many require subscriptions. De Montfort University, like many other universities, has built up an openurl resolver listing then different journals their members can access for free. It is possible to get this service ‘plugged in’ to Mendeley.

  • Go to Mendeley and log in;
  • Go to the ‘My Account’ section;
  • Select ‘Account details’ from the drop-down menu;
  • Go to the ‘Sharing/Importing’ tab;
  • In the ‘Edit Library Access Lonks’ section, click on the ‘Choose library from list’ button;
  • – In my case other settings prompted Mendeley to suggest ‘library links’ relating the De Montfort University. If your institution is not listed, you may need to use the ‘Add library manually’ button;

You should now find that when looking at a citation page on Mendeley there is a drop down-menu ‘Find this paper at:’ which lists your institution’s openurl resolver. Selecting that should point you at the links to the journal article that you are able to use without charge.

Posted in Library resources, Mashed Libraries, OpenURL, Social media | Comments Off on Tuning Mendeley to find electronic journal articles

Discounts for students from @MyUNiDAYS

MyUNiDays is a student discount website offering ‘perks’ from a number of online retailers.

There are new and well-known retailers signing up to the growing range of stores represented with deals of 10-20% off for students. You can prove that you are a student using the same Single Sign On details as you would for Blackboard, MyDMU and university email.

If you are interested in exploring this:

  1. Click on the ‘Join‘ button;
  2. You can use any of your email addresses and set your own password;
  3. Start typing ‘De Montf..’ and select ‘De Montfort University (Leicester)’ as your institution and click on the verify button;
  4. A new window opens with the Single Sign On to Library Resources screen. Type in the username and password that you would usually use for logging in at DMU;
  5. If all has gone well you should now be ‘verified’ and can start browsing the different sites associated with MyUNiDAYS.

You can follow MyUNiDAYS on Twitter for updates and new ‘perks’.

Posted in Authentication | Comments Off on Discounts for students from @MyUNiDAYS