Jump to content

Recommended Posts

Posted
I'd go down the web hosting route for library systems. Our school is becoming an academy so it'll be nice to see what resources are in the other schools within our MAT.

My wife is onsite at one of her schools doing weeding during half-term on their web-hosted library. Each delete was taking 20-30 seconds.

She tried using another machine to try to speed this up (my daughter is there helping too) but it failed to connect. 40 mins on the phone with support and they talked her through installing something or other which has now crashed that machine (Kaspersky AV claiming Malware found) and they just gave up and said to contact the school IT department to fix it.

I drove there and tried the Barcode multiselect to delete multiple books at once. The screen didn't match the documentation and failed to delete - just kept hanging for 30 seconds and then doing nothing.

After trying this for a while, I restarted the app and logged in again. It seems to work after that - still very slow but deleting 10 books at a time helps.

 

Seeing resources in other libraries is a good reason to look at web-hosted options but just make sure that whatever system you choose performs acceptably well on your own library otherwise that benefit will be outweighed by the drawbacks.

Posted
Since others are talking price, we pay £500 per year for this, locally hosted. Web hosting is the same amount again I think.

Sorry, does that mean Web hosting is also £500pa or £1000pa?

Posted

SIMS: there are lots of "default reports", but all for specific purposes. You could probably find a native report which includes all the fields you need but it would include lots of other data you don't need; for an on-site system, this wouldn't be so bad but there's no way we would send you all that data to a hosted system. You'd be far better either writing the SIMS report yourself or telling customers what fields to include in the CSV and have them make their own report; to be fair, most services we use just list the fields and leave us to provide the CSVs.

 

Photos: honestly not sure how photos are extracted, but probably just as JPGs named as the students' admission numbers.

 

User import: possibly just your phrasing, but I wouldn't want a system which required IT to "shuffle existing groups". I want to fire a CSV which has student names and year groups in at the system. The system should compare that CSV against existing users, update any who were already there, create any new ones in the CSV and offer to delete any who are not the CSV. The thing to bear in mind is your system will not be the only one IT are having to maintain (I count 11 systems here for students, and another 4 for staff) so the process should be as soft-touch as possible (hence my preference for automation). On that note, please give the school choice of username and password format, so they can keep consistency with their other systems.

 

Our system: we're using Access-It

 

Price: £500 pa for the local system, and another £500 or so for hosting, so £1000 total for hosted.

 

If your system was better but more expensive, we would consider moving; it is unlikely, however, we'd move to an inferior product in the interest of saving money.

Posted

Our system: we're using Access-It

 

Price: £500 pa for the local system, and another £500 or so for hosting, so £1000 total for hosted.

 

Sorry to diverge, but how are you finding Access-IT?

Posted
Sorry to diverge, but how are you finding Access-IT?

 

It works. User creation is automated. Cataloguing works quickly. Students are emailed automatically about overdue books (a way of emailing parents about very overdue books would be a nice addition). Support can be a bit slow and sometimes tend to blame "local network problems" until you can prove otherwise. Librarians seem happy enough, I rarely touch it so I'm happy too.

 

Not installed v9 yet, so can't comment on that.

  • Thanks 1
Posted
offer to delete any who are not the CSV

 

Thinking about it, it would need to offer to delete leavers rather than do it automatically in case someone has left without returning an item.

  • 2 weeks later...
  • 4 weeks later...
Posted

OK, so its about a month since my last post so here is an update..,

 

I have the following proofs-of-concept now working:

- Complete, structured MS Sql database

- Sample database of 10,100+ items (an existing database imported)

- Grid showing all Items and all columns in the library

- Can scroll through all items on a large screen using scrollbar with zero delay - none of that page-at-a-time-then-wait rubbish

- Many filters - some directly on the grid, some custom ones - all work virtually instantaneously.

- Custom annotations - e.g. Borrower column shows red stripe and tooltip says how many days that item is overdue.

- Barcode scanner support

- I bought a cordless one from Amazon - reasonable price - typical features.

- Most of these work as a virtual keyboard and when scanned pretend the digits plus Enter have been typed.

- This is fine unless you happen to be in the wrong field where upon you might overwrite data!

- I spent a lot of time researching low-level USB hardware/software interfaces (not normally my thing) so that I can detect and intercept when a scan has taken place and convert it to an event such that the app can deal with it properly. Still needs a bit of work identifying the scanner in the first place because my machine thinks I have 7 USB keyboards attached!)

- Dashboard

- Identical to the MicroLibrarian one including the pie-chart and percentage-filled images (I even bought the same icons)

- Filtered Lookups

- If you have several thousand people in your database (my test one has 4,610), and you lookup in the Author field, for example, how do you decide what to show? You probably don't want to show the schoolkids names or teachers, so you might only want to show Persons Tagged as Authors. But on, this occasion, you do actually want to find a schoolkid's name. My solution is Named Filters that are changeable directly on the drop-down:

- e.g. Authors Only; External People; Internal People; People; Show All - no limit on the number of filters that can be defined

- To be added on lookup : ability to create Person or Organization on the fly/Force refresh from database

- AlertChecker

- Runs background checks against your data and shows a list of items failing the check

- Any number of checkers can be included

- Will show a list of the items and allow double-click to open the item

- Some checks allow you to select and automatically fix the items

- Special feature allows you to mark an item as Valid and it won't be reported again.

e.g. a Check suggests renaming a Title to something with different case but you disagree. Just mark it as correct and you won't see it mentioned again

- Labels

- Support for Avery L7651 65 labels/sheet (actually A4 Labels for laser, inket & copier, best price compatibles) - mm perfect.

- Other sizes to follow

- Accession labels - bar code and number

- Class labels - Dewey number and wrapped title

- Borrower labels - bar code of logon ID and name of Borrower, one sheet per form - for younger primary school classes

- Bibliographic information (for enjay and Steve21 in particular !)

- Given an ISBN, will connect simultaneously to several sources and parse relevant data

- Current sources are Google Books, Open Library, Library Thing, Amazon UK, British Library, Library of Congress

- Includes book covers where available (all except BL and LoC)

- For the technical minded - I have support for XML, JSON, MARC sources

 

Is there anyone in this forum interested in looking at (screenshots initially, PC app in a few weeks) and helping to further spec. the app?

This is your chance to describe exactly what you want from a Library system. All you would need to to install the free SQL Server Express on your local PC.

If so, please feel free to contact me with a PM.

Posted

Here is a screenshot of the biblio lookup form...

 

LookupISBN.png

 

The plan is that the left column will hold the current values for the resource. Double-clicking on a cell over to the right (or maybe each cell will get a button*) will copy its value over to the left column.

 

Save will then apply those left-column values into the Resource. There will be a bit of extra work setting the Author or Publisher for example, since these values are only strings. So we need to be able to select an existing "Bloomsbury Publishing Plc" rather than creating a new entry.

 

Here is a screenshot of the Alertcheck screen....

 

AlertChecks.png

 

(I have some big plans for the Title check. I researched some code (OpenNLP) that should help infer context and be more authoritative about which words should be capitalised)

 

Here is a screenshot of the Grid and popup Filter Panel....

 

Grid.png

 

The left two columns are fixed so the horizontal scrollbar will show more columns but leave the Title always displayed.

I plan to add column sets to the Filter Panel (or maybe another popup panel*) so that you can show/hide the Borrower/Due Date/OD Days at once.

 

Tomorrow, I'll swap the Borrowers names randomly so I can upload screenshots without having to blur the names.

 

* These are the kind of thoughts I have that I need direction on! Otherwise I end up coding stuff that noone wants.

 

Cheers

Simon

  • 1 month later...
Posted (edited)

Latest update:-

Bibliograph Lookup

The biblio lookup form has been updated:-

001.png

- The form is now split into Left and Right. Left shows the values that you want to use and Right is a scrolling list of biblio sources

- Note the arrows on each item on the right. They allow you to choose which bits of text you want to use and the values are copied to the left.

- or you can click on the arrow on the first row and copy all available values from that Source.

- or you can use the Get Best Values button to select the best available value from all Sources.

- All values shown are simple strings but that is not what we want in our library so we need to match those strings against existing items (or create new ones)

- The coloured dots on the Left values show whether there is a direct match in the system (green) or whether there is a similar match (yellow) for you to choose.

- When you click save, you will be shown a further form to showing for each of those imported string values a dropdown of equal or similar values in the library for you to choose.

002.png

- In this case, the classification of 823.914 doesn't exist so 823 is nearest. However, the dropdown also has a Create... option so, if you wish, you can create 823.914 on the fly and use it immediately.

- Creating Authors, Series names, Publishers will also have a Create... option

I can't think of anything more to do on this. It works very well.

 

Word Cloud

I have created a simple word cloud as a proof of concept - takes list of strings and their usage count and draw a cloud:-

003.png

- This will be available to show clouds of Keywords, Searches (whole library or per class), or anything we can think of.

 

Classification Keywords

- All Classifications can have Keywords associated with them. This will help Searching where individual resources have not been keyworded yet.

004.png

Item Links

This is not complete yet, but I have made a good start. The idea is that a 'thing' in the database can show how it is linked to other 'things' in the database, allowing you to jump between them. Each 'thing' has its own summary form.

For example, here is a summary form for a book resource:-

006.png

Clicking the link button at the bottom left, shows this:-

005.png

This makes it very easy to find similar resources. for example, clicking on the "Horrible Science" Series and its link button will then show this:-

 

007.png

 

Then clicking on those links (one or multiple) will then show that book's summary and so on.

When you find what you are looking for, click on its Open button.

 

The same concept applies to People, Organizations, Classifications, Keywords etc.

Edited by simmotech
Images went missing
Posted

I have now come up with a good way of anonymizing the personal data within my test database so it is now possible to let anyone interested to try out and help further spec the application.

 

I shortly plan to publish an Alpha version of the software on my website so you guys can have a play with it.

This will be a One-Click .NET publication so you will get the latest published update.

(If you are paranoid that I am trying to infect your computer, just run it in a virtual machine sandbox).

The app can be uninstalled at any time using the normal Uninstall in Windows.

 

Important! There is absolutely no obligation on your part to buy the final release should you try out the Alpha version!

This is purely a way of allowing me to get input on required specifications which will help the application sell itself later.

(and also bug-finding of course!)

 

You will need a version of MS SQL Express on your testing machine and preferably the SQL Tools as well.

The App will look for a database on the current machine and not finding one, will show the following screen:-

 

008.png

 

You can either create a completely new database using the top arrow-button; or copy the same sample test database (10,000+ items) I am using using the bottom-arrow button. I suggest you use the latter - it will Restore the database in less than a second and give you a good number of objects to play with.

 

(Please note however, that I consider the test database poor quality overall as it was imported from a Junior Library database and is full of mispellings, duplications, and poor choices of reference data etc. It is useful to me however as it will allow me to develop maintenance tools to 'cleanup' imported databases - see the "Series Name" maintenance where Merge button allows multiple items to be merged into one)

 

Using SQL Management tools, you can delete the database you created at any time and you will see the above screen the next time you run the app.

 

Please feel to PM me if you need more information.

  • 3 weeks later...
Posted

Update: Not a single person interested in having a play! Oh well, plod on...

 

As it happens whilst only half-way through development, I now have a customer!

It is a Primary School that is going to give Junior Librarian the heave-ho and so I now need a minimal working version by mid-April which is when their notice period to MLS will be up.

So the first release will be necessarily biased towards supporting Primary School scenarios

 

I have access to a couple of hosted Junior Library databases and have worked out how to extract data to recreate the library in my software.

Unfortunately MLS have problems with CSV exports (corrupt), MARC21 (invalid) and XML exports (can crash if >~36K records).

But my first school has a relatively small library so I have a full set of data to play with.

Also, I am talking to MLS Support about those and other issues I can easily recreate so they can get them fixed (even after all I've said about them, I still feel a bit guilty about writing rival software to pinch their customers)

 

My question for you today is about Media Types. Because it is a simple text field in MLS, everyone and their dogs type in seemingly endless variations and then wonder why they get 'incomplete' results when they search.

e.g. These are just the new ones just from my Primary School that I will have to map to my existing Media Types:

big bookbook (details unknown)book (details unknown);audio cassettemultiple copy packnovelty bookohp transparenciesppaperback (a format)paperback (c format)paperback / softbackpaperback;audiobook on cassettepaperback;audiobook on cdpaperback;cd-rompicture bookpicture booksspiral / comb boundspiral bound

 

My software will have dropdown selections and I have some defined at the moment (ones with * may be questionable)

Book

Hardcover *

Hardcover with dust jacket *

Paperback

Book and tape set *

Novelty Book

Board Book

Leather/fine binding *

Mixed Media *

Map

Picture Book *

Audio Cassette

CD-ROM

DVD

Blu-Ray

E-Book

Software

 

For each Resource Type, the dropdown will show a certain subset of the above.

 

My problem is that I am not a librarian so I don't know how far to go in supporting Media Types.

Does anyone really need to catalogue to the extent they know whether a book is "big" or is "spiral bound".

 

Let me know what you think. Thanks.

Posted
My question for you today is about Media Types. Because it is a simple text field in MLS, everyone and their dogs type in seemingly endless variations and then wonder why they get 'incomplete' results when they search.

e.g. These are just the new ones just from my Primary School that I will have to map to my existing Media Types:

big bookbook (details unknown)book (details unknown);audio cassettemultiple copy packnovelty bookohp transparenciesppaperback (a format)paperback (c format)paperback / softbackpaperback;audiobook on cassettepaperback;audiobook on cdpaperback;cd-rompicture bookpicture booksspiral / comb boundspiral bound

 

...

 

My problem is that I am not a librarian so I don't know how far to go in supporting Media Types.

Does anyone really need to catalogue to the extent they know whether a book is "big" or is "spiral bound".

 

I believe you are referring to a GMD.

Posted
I guess it should be a customisable predefined list, so people can add extra values

That is the easiest way but not necessarily the best way for future usage.

 

What I ideally want is to find out those 'extra values' and include them into the system (ie they are 'well-known', have their own ID and can be programmed against).

 

That's not to say they are useful to everyone all the time - I can provide an option that allows which 'well-known' values are allowed in that specific library.

Posted (edited)
I believe you are referring to a GMD.

I had to look this up and there are dozens of definitions of GMD - to which were you referring?

Edited by simmotech
Posted

Well, it seems I have my first customer!

 

A Primary school is refusing to renew their upcoming MLS Junior Librarian subscription because their cloud-hosted system is down more than it is up and very, very slow at the best of times and too expensive.

 

So my development is now focused on providing solely what they need for that initial release mid-April:-

- I have already (mostly*) extracted their data from the cloud in XML and have an Importer to read it all in.

- I just ordered a cheap and cheerful webcam from Amazon so that that initial release will support images of the kids. (* can't export photos from JL easily that I can see).

- For now, circulation will be hosted by class - ie a single screen showing everyone in a given class and allowing the Librarian to return their books and let them choose new ones.

 

This simple requirement means I should be able to get what they need working by mid-April and I can write more the more complicated stuff later.

  • Thanks 1
Posted

Free School Meals, Pupil Premium, Special Educational Needs;

Language; Ethnicity

 

Is there any reason a school library would make use of the above fields?

Posted

Free School Meals - not relevant so not included.

Pupil Premium - although my wife points out schools with Pupil Premium pupils enables them to pay for her very reasonably priced services, I can't see how it is relevant in running a library so not included.

 

Special Educational Needs - JL just has a checkbox for this but it sounds like a whole range could be included.

Language - Not sure what it means - Surely English is what is being taught, so does it mean the main language of their families?

Ethnicity - There could be a shitstorm here if my dropdown doesn't cover a particular ethnicity - https://en.wikipedia.org/wiki/Self_defined_ethnicity - would rather not go there!**

 

All 'Persons' in my database can have Tags associated with them. Some of these Tags are built-in but others can be added.

I think it would be better if I left Special Education Needs/Language/Ethnicity out for now and let librarians add custom Tags to more clearly define what they want to identify.

 

 

** Already had a telling off from my 14yr daughter because the boy/girl default Avatars I had were too light-skinned' - she told me they should be Yellow by default like Emojis in iPhones.

And she was right! UTS #51: Unicode Emoji so I fixed that up.

 

Avatars.png

 

 

My cheapo test webcam works well but the controls I have are extracting a single-frame from the live video sequence and I haven't worked out how or even if it is possible to get a proper snapshot at photo resolution that would work on all webcams.

Also, being a video recording, the images are landscape where we would need a portrait image. I may have to split this into two parts - one to get an image and then one to crop a certain size out of it.

Luckily, my first customer didn't use photographs so I can spend a bit more time on this later.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...