dhicks

I tried ExpanDrive, with much the same results as you - it doesn't mount the drive as a "regular" file system (there was a specific term but I can't remember it), so Veeam doesn't see it. I did think of trying with rclone, which I've used to mount various cloud-based file system on Linux, but I've not tried it with Windows yet.

Depending on what you're filming and what sort of run-time you need, GoPros can get quite hot in use - or at least previous models did, the newest generation might be better. I've tried plugging them in to an external power supply and leaving them to run for 2 hours to record a play or similar and they tend to overheat. Might be worth considering the Padcaster if you already have / use iPads - turns an iPad into a camera, complete with wider lense and external mic.

We're generally using Google Meet to live-stream video (which does have a record option), we also have a WeVideo subscription. Both the camera input and the mic input are seen as standard no-drivers-needed USB camera / audio inputs by both Windows and ChromeOS (haven't tried on MacOS yet, but should be the same), so just make sure you have the relevant input selected before you start. If you want to go for a camcorder (quite cheap, gives the person being filmed a swivel screen to check position), we use Panasonics with HDMI out and a 5V power input, which means you can power them from a standard USB battery pack. This means you can stand the camera in a busy room without cable clutter everywhere, which is very handy. Personally, I prefer the stackable single-pole mic style stands with a thread adaptor to take a camera than a traditional tripod stand, there's less for people to trip over. I've just bought (but not tested yet) a £20 HDMI-to-USB adaptor from Amazon, if I get a chance to try it out soon I'll let you know how I get on. Otherwise, there's the BlackMagic Atem Mini if you want to be able to switch between 4 different HDMI channels and have a standard USB camera input. A USB sound interface will generally give you a output for zero-latency monitoring which you can plug a set of headphones in to hear yourself speaking. I've set ours up with a 4-way headphones splitter so both the person talking and myself can monitor the sound. Again, this is a 5V-powered device so I can power it from a battery pack or the same computer the sound interface is plugged in to, reducing hum on the headphones circuit.

We've got a Logitech PTZ Pro - plus motorised pan / tilt and zoom, all operable from a simple remote control. Plugs in to a PC / Chromebook as a standard USB camera, no extra drivers needed. It produces a nice image, btter than your average webcam, and can zoom out to a good wide shot and in to focus on the speaker. That camera is currently around £750 on Amazon, which might be a bit expensive - we plan to use ours come Christmas, etc, for streaming plays and so forth. I've got an external USB mic interface for use with the above, £30, which then allows you to attach a proper condenser mic, complete with 48v phantom power, for sound. Again, plugs in to any PC / Chromebook and is seen as a standard USB micr, no driver needed. We're going to try a cheap (£30) shotgun mic and see how we get on. For a "pro" effect on videos, sorting the sound might be as important as the actual video - recording with a decent mic and in a not-too-echoy room is probably going to help.

Agreed, and it's interesting just how little there is to Google Classroom - it's basically a joining table sat between the list of users and various Docs / Sheets / Meet / etc resources. It turns out that that's pretty much all the functionality that's needed. Microsoft's .DOCX format - actually surprisingly open and easy to work with. Turns out you can rename a file to .zip, extract the files, do a search-and-replace inside the XML, then re-zip the whole lot back up again, rename it .DOCX and it will be a valid file. I didn't have to read a specification or import a library or, actually, even write a script to do that. Both Google Docs and Office 365's web-based versions have more limited functionality when it comes to editing Word documents (Docs doesn't let you float images in front of txt, for instance), but that seems to be a deliberate choice to limit the amount of mangled formatting the average user can create. We have a couple of files on our Windows shared drive that Word now can't render as PDFs, quite possibly because so many people have added so many different bits of formatting to the document that the PDF engine gets stuck in a loop trying to render it.

We're just starting to do the opposite - sync a Google Drive folder or Shared Drive to a local file server, then back that up with Veeam. Staff or pupils can access their files via Google Drive or, if needed, install the Drive File Stream client. We have this working for our "media" storage drive, 3.5 TB(!) of images and videos - staff seem to be happy to access via ChromeOS / web browsers, we haven't had much use through the File Stream client yet.

For the Raspberry Pis used in the curriculum, we have VNC enabled (the built-in RealVNC, I think), plus the somewhat-confusingly-named NoVNC, which puts a web interface in front of VNC so you can access it via your web browser. This lets us plug in the Raspberry Pis to power and connect by simply opening a web browser - handy as we have iMacs in our ICT lab, so no monitors with HDMI in to plug a Pi into.

In theory, I have a nice home working setup with an original IBM Model M keyboard and Microsoft Intellimouse, both connected to a NUC plugged in to a 32" TV / monitor. However, my girlfriend has been using that for the last 6 months in the front room while I've been in the bedroom on an Ikea console table with my laptop. For that, I have a Logitech M-UV55A mouse, with the main distinguishing feature of having a not-too-long cable that doesn't get in the way.

Just seen this advertised: https://thepihut.com/collections/featured-products/products/argon-one-raspberry-pi-4-case

I've got a Raspberry Pi 4 in one of those cases running on my desktop here in front of me - I can't hear the fans at all, they are pretty much silent. In the picture in the link above you can see a power cable coming from the fans to the Pi's pin interface, this actually tucks in to the side of the pins quite nicely, so there isn't a bit of wire dangling off the Pi to get snagged on things. The fans are just bare, there's nothing nothing to stop you / small children / pets / random passers-by poking fingers / noses / pencils / cheese sticks / etc into them - I've not tried it myself, I doubt they are particularly dangerous, but worth pointing out.

Our main line went down (fiber line provided by Exponential-e) but our secondary line was okay (standard BT fiber-to-the-cabinet ADSL service).

I came accross this service a few weeks ago: https://tunnelto.dev/ Allows you to tunnel a port through a firewall, complete with HTTPS managed for you - so you go to https://someurl.tunnelto.dev, your traffic gets forwarded to your server behind your firewall. Cross-platform, seems to work nicely on Windows, you just need to install it using NSSM if you want it to act as a service.

Ah, that's interesting / useful to know - I've not had experience with either of those, it's interesting they have different approaches. I think I prefer the option to pay a (reasonable) extra fee for limited write-back functionality rather than have to wait for someone to approve any code I'm writing.

API compatibility might be an interesting place to start if you were looking at creating your own system - if you looked at something like Clever / Wonde and worked backwards to produce a minimal system capable of providing the data to those external systems, then only added features and data as need to support further external tools, the aim being to produce the minimum system needed to support a set of external tools that actually implement the features you want. We've found several tools / services that use the iSAMS API directly, too, so you could perhaps create something API compatible with that.

How is the site's login process implemented - can you submit a username / password from your own login page? If so, you could set everyone's password to the same but (to them) secret password and have users login through your own login page where you could provide integration with GSuite / O365 / etc.

There are some tools / systems available that give a common API across different several MIS systems, such as Clever (which is mainly US-based) or Wonde in the UK. You could aim to build the functionality you want on top of one of those systems. How much experience of software development do you have - is this something you're considering doing yourself, or something that you would hire someone to do? I would suggest that starting an entire new product, both server side and front-end, from scratch would be a sizeable undertaking. If it's any help, I find iSAMS' API pretty reasonable, although it doesn't cover every part of their functionality and is generally read-only. If you could maybe elaborate a little bit on the kind of thing you need your MIS to include that might give a better idea of the kind of solution that might be appropriate.

Does the wireless reconnect after a restart? If so, is it practical to have the machine restart to do a reconnect? That way, you could have a script running that monitored a ping command or similar and did a restart if it found network issues.

One option might be to install ChromeOS on the old Macs - Cloud Ready should install okay, and you get to carry on using your nice Mac hardware. Soundtrap has grown in to a very capable-looking solution, complete with features such as Autotune and so on. The great part is that it should work from pretty much any computer with a browser, and pupils can use it at home. If you're going to spend money anywhere (as ICT Direct point out), spend it on some USB audio interfaces, mics and headphones. You can even get USB Midi keyboards (about £50) that can work with APIs available in the browser, so you don't even need to get a separate Midi interface.

Macs generally keep their resale value quite well - 2013 iMacs are currently going for around £300 - £400 on eBay. You might find someone to buy the full set off you, or possibly someone that will take existing equipment in part-exchange.

The 19" rackmount standard is used for music studio equipment, so you might find something suitible at Gear4Music or Studio Spares. The thing to search for might be "stage box", we have a couple of 2U units holding amplifiers and mixers on our stage.

Storage Spaces Direct? I can't remember which bit of functionality that is - I know we configured the Windows server to have a local "pool" of storage with an underlying tier of larger disks (several 8TB 3.5" drives) with a faster tier above that (two 1TB SSDs), so most writes (and a lot of the reads) will happen on the fast SSD drives. We don't have any real-time failover for VMs set up - both our email and MIS are now web-based, we're starting to move the file servers the same way, so there really isn't going to be that much left running locally.

Six years on - we stuck with DRBD mirrored volumes on XenServer for our (school / work) production server setup, and although it's now looking dated we're still only just sorting out moving the VMS and storage off, so it's lasted very well! For the new server setup we've gone with basic Windows Server 2019 with Storage Spaces - we needed to run Windows Server VMs, so the licensing needed to be in place anyway, and we get vendor-supported hardware. It's getting less relevant for the average school / business to host their own server hardware - even our new file server is probably going to mostly be moved for backups once we get the working file shares on to Google Drive. In terms of the kind of computational / storage problem that Ceph is built to handle (if I understand correctly, AWS Lambda-style jobs with VMs that start up, connect to a block-based storage pool, then quit when done), I'm having trouble finding tasks suitible for that kind of computing model. Day-to-day tasks still tend to best be solvable (as far as I can see) with a program that gathers inputs (generally, files from a traditional file system), does some processing, then produces some outputs (again, often back out to a file system).

Indeed - that's the problem I'm trying to work around. It does seem to pass login details into Chrome, though, hence my plan as detailed above. If I can get a not-too-clunky AutoIt script or similar put together I'll post back here somewhere.

If that cache is set to 10Kb, that might be a bit small. Do you mean you're having to log in to the File Streaming client every time you log in to Windows or just every time you log in to a machine for the first time? I've been trying to figure out how to get the Google Windows login component integrated with Google File Streaming, so the user doesn't have to set up the file streaming client separatly, they just need to log in once. I think it might be possible to force the File Streaming client to authenticate via Chrome, then have an AutoIt script or similar start the File Streaming client and authorise via Chrome. If I can figure it out properly, I think that might even work to get the File Streaming client going before Windows File Explorer starts up (if I replace the default Windows shell with the AutoIt script), so a user's Documents and Desktop folders can be redirected to Google Drive.

Just did a quick check - turns out that the File Streaming client caps cache usage at "20% of free harddrive space": https://support.google.com/a/answer/7644837?hl=enbn However, I don't know if that winds up being 20% per user, which in a shared environment could get large quite quickly. You can set a hard limit in kilobytes yourself. I suspect File Stream might well do some pre-emptive caching of your most popular files or the files it thinks you might need next.