So... it's been a while. Some life and health stuff interrupted my ability to work for quite some time, so I wasn't able to make progress on any of my projects at all.

At last, I've been gearing up to get back into the hot seat - you may have noticed issue publishing has resumed recently - and while I haven't quite tackled my somewhat overwhelming two years' worth of email yet (apologies if you've emailed in and not received a response), I have been re-acquainting myself with all the various parts of the mu:zines workflow, and getting back into the various projects I had to abandon for a while.

Let's summarise the recent activity.

Scanning

The first thing I got back into a few months ago was tackling my scan queue, which had around 100 issues remaining.

After a few years of waiting for VueScan, my long-time scanning software of choice, to finally add macOS ImageCapture support (which the developers had said was something they were working on), I checked in with them again, as I did regularly, and they finally admitted that it wasn't happening.

Without this support, I could not use VueScan on my current Mac. Instead, I have to dig out my old, often-not-booting and running-on-fumes Intel Mac and scan inside a Windows VM, which was clunky and a bit of a pain. So I decided to leave VueScan behind and try to find a new scan workflow to take its place.


After a bit of investigation, turning a problem into an opportunity, and building up some new custom scan monitor tools, I now have an entirely new and better scan workflow that I can run on my regular machine. It removes a lot of the friction from scanning, improves various parts of the process, and makes it much easier to catch errors and so on.

I won't go into the details of all the ways the improved tools help the scanning process right now, otherwise this blog will become a monster(!)

I've got through 60+ issues, including finishing scanning all of The Mix, and a bunch of fat International Musician issues, bringing my scan queue down to 25 or so massive IM behemoths left to go. I'll be getting through them, and probably scanning more stuff not yet on the site, over the coming weeks.

Improvements

I started going through and fixing some pain points in my admin tools: things that can bog down the process a little, minor but frequent annoyances, and things I've always wanted but hadn't got around to implementing. I then started doing the same for things on the public-facing site that had bugged me for a long time.

Search

The first was the search box, which used to have a pop-up behaviour that I disliked, but could never previously find a solution for. I wanted it to stay in place and simply expand when clicked.

I implemented that, and also fixed a bunch of responsive issues with the header/navigation, particularly on mobile devices. It's much better now.

Tables

Tables have been a pain point for some time. Small tables with, for example, gear specs are not a problem, but large, often multiple page-spanning tables with lots of data just aren't a good fit for a narrow-ish column article format. I had implemented a bunch of workarounds for this in the past - making the text small, putting tables into wide blocks that run underneath the sidebar, and so on - but there were always situations where none of these were great solutions.

So now we have a "Spreadsheet View". All tables that would benefit from it now have an "Open this table in spreadsheet view..." button, which opens the table in a new browser tab and makes it much easier to read. You can re-size and re-order columns, use your browser search, etc, and it's generally a much better experience if you're interested in the details of some of these tables.

You can see how it works with some of these example articles:

- Mixdown (SEP Feb 86)
- Tape Machine Guide (SEP Feb 86)

I've also gone back and added this handling for the articles in the archive with the worst table offenders.

Full Scans View

Another long-standing pain point was the viewer used to browse the full page scans. It wasn't great, or very flexible. It was certainly fast to navigate through a magazine this way, using cursor keys to flip pages, but it wasn't a good way to actually read the contents for those who wanted to. The sizing was inflexible and the images too small to really read any details.

So it had been on my list to re-do from scratch for some time. Before I had my health issues, I had done the work to investigate and prototype a new fullview system, and although I hadn't implemented everything mu:zines needed, it certainly behaved well enough to serve as a good foundation.

Bringing that prototype forward, integrating it into the site, improving performance, and adding the remaining features has been my task for the last week. It turned out to be somewhat more extensive work than I was expecting, but I'm much happier with the new system.

It's fully responsive, works across all devices and display sizes, lets you quickly browse through pages using cursor keys or the mouse, supports magazine spread layouts in addition to article page views, collapses down to a single page view for constrained size devices, has a zoom mode that lets you zoom in and drag around to read smaller details, and lets you rotate pages (for those pesky adverts printed sideways, etc).

It's also now used when doing advert searches, so you can better inspect advert scans in the same manner.

It's a much better way to browse the actual scanned magazines, and is fun in full screen mode with a large monitor.

Scan Resolution

To support the ability to better browse and examine the page scans, and read the small details, it's no good if the page scans aren't high enough quality to do that. So I've re-output all the site page scans - over 50,000 page images - at a higher resolution.

Magazines like Making Music, which had A3-size pages and were more or less unreadable with the previous viewer, I've outputted even larger, so it's still possible to read the smallest details. While webserver space and internet performance constraints mean I can't put super high-resolution pages online right now, this is much better than what we had previously. I will be outputting all future issues at this new larger size going forward.

Other Bits and Pieces

Another thing on my list for some time was that some of the earlier scans I did for this project weren't great. I'll probably end up re-scanning some of these issues if I can, but for now I went through and re-cropped about two years of MT issues that were pretty rough-looking by my current standards.

I've also been going through my issue/bug tracker and fixing a few more niggling site issues, making improvements, and will continue to do so. There are also other features going on in the background that are not yet finished and ready for prime time.

Notable Mentions

Another big task that I started a while ago, but haven't explicitly mentioned, is handling image credits.

When I originally started the CMS (some twenty years ago now!) and was assessing how to get a magazine into it, I chose at the time not to explicitly handle image credits, for a bunch of practical reasons. I figured that, as we have the scans available, the credits would still be viewable anyway, so it was an efficiency tradeoff I made back then.

However, over time I really started to feel like I should be handling these better. I do from time to time get image licensing requests, and do my best to pass them on to the owner of the images in question. It still didn't really make sense to build out a complete system for handling image licensing requests, but I had some discussions about credits with one of the photographers who contacted me, and it was enough of a prompt to overhaul this and handle them better. So now I handle and credit images where there are explicit credits in an article.

I also therefore had to go back through the archive and start to tag the existing article images. This is a sizeable task, as there are tens of thousands of images, and the process literally involves checking every single article that has images, seeing if there are explicit image credits, and re-tagging them accordingly.

I have been working through these systematically, and have to date done thousands of images, but it's a process I'm slowly working through.

Moving Forward...

So - what next? Well, for me the key thing was just to get back in the flow of moving the project forward. I'm continuing to work on new issues, scanning outstanding issues, improving the site and my workflows, and implementing things on the wishlist to continue to make this site a useful, and somewhat unique, resource.

And maybe even some new blog posts, if you can stand the excitement... ;)

Thanks for your patience in bearing with me!