dhicks

As it turns out, I wound up buying a Kodak SP360 4K - currently stocked, it would seem, by Jessops, but not Amazon. I bought the single camera, not the two-camera setup, so I have a device that can see around 235 degrees. It suits what I want to use it for, which is filming live events (comedy shows, school plays, etc) in a discreet fashion - with a wide field of view you can set up a camera on a monopod near the front of a stage and capture all the on-stage action for later editing without having to be standing there getting in the way with a camera. The hardware compares well with my GoPro - the SP360 is small, action-camera sized, slightly smaller than the GoPro. I used it to film a friend's comedy show, running continuously for 3 hours or so (I had a 128GB micro SD card for storage raw - footage seems to be recorded with a bitrate of around 60kbps, so make sure your SD card can handle that). I ran it plugged in to an external USB battery, which seemed to work well - after three hours the battery on the external battery still had plenty of charge, and the battery in the camera itself still reckoned it was full, so with a full external battery your limitation is probably going to be the memory size (128GB gets your around 4 hours recording time). The camera does get quite warm, but not as warm as the GoPro, which got nearly too hot to touch when running off an external power supply for a couple of hours. The software available from Kodak (well, JK imaging) for the camera is actually okay once you get the hang of the interface. You can trim clips to length and convert to YouTube-uploadable equirectangular format, and that's about it. The camera itself records in 4GB chunks, so you'll need to get something to stick MP4 files together. The software handles those combined files just fine, though, so it's not a huge issue.

We've been using a Chord 4-way rackmount unit for the past 2 and a bit years: http://www.edugeek.net/forums/av-multimedia-related/113765-rackmount-wireless-microphone-set.html Available for around £400 on Amazon, comes complete with four mics and an all-in-one mixer that lets you output either individual mic channels or a single mixed output from all four. Can be bought as a set with either hand-held or headset mics, plus you can get individual mics / headsets (Premier Farnell had reasonable prices the other week when I looked). We actually bought two sets, one of handheld mics and one of headsets, so we can swap between the two. We often use the above with a small, portable mini-amp: https://www.studiospares.com/Headphones-and-Speakers/Portable-PA/Studiospares-Trojan-iPA800---Mic-Stand---Bag_250212.htm That lasts well on battery power, although you can plug it in, and has a proper output that you can connect up a second speaker to if you need more coverage. Again, nothing fancy, but it provides an input for both your microphones and a PC / iPhone or similar (it has both an iPhone cradle on top and phono inputs). It'll stand on a standard speaker stand, and you can put the rackmount wireless mic receiver in a 2U stage box and have a nice compact, portable, plug-in-and-away-you-go setup for somewhere under £600. Chord now do a six-way mic set which I'm tempted to get but haven't actually tried yet. The Chord systems are probably at the lower end of the audio quality scale, but I find they are excellent value for money for the perfectly servicable equipment they produce. Do be wary of some of the even cheaper units available on Amazon and eBay - some of them use frequencies that aren't legal for us to use in the UK.

We haven't gotten around to configuring any quality of service settings for our 60-odd phones yet, they've been working just fine for the past 2 years without being on a VLAN or having QoS settings set up. The plan is to use software defined networking to do the QoS part anyway, not VLANs, but there's always been something else to set up first. I should perhapse add that we have a 10 gigabit fibre backbone, so we probably just have plenty of bandwidth to spare - we probably should set up QoS eventually. The same is likly to hold for our external lines. At the moment we have 4 external phone lines provided by ISDN. As discussed previously in this thread, we can replace those with SIP connections of some sort. In theory, you need garunteed quality-of-service settings for your external SIP traffic. In practice, we have a 200mbps fibre-to-the-premisis connection and there's currently plenty of bandwidth to spare (although it would be sensible to have our gateway prioratise SIP traffic). Audio data actually takes up relativly little bandwidth these days compared to all the other stuff we use the network for.

I was under the impression that a SIP "trunk" was simply a server sat somewhere, hooked up to a bunch of phone lines, waiting for connections from SIP clients. Do ISPs activly stop users connecting to SIP servers in some way, or does "not supported" just mean we can't complain at the ISP if bandwidth / latency to a particular SIP providor proves poor?

As with what seems to be most of the posts above, we installed our own PBX and some IP-based handsets. In our case an iQ PBX and Yealink handsets. The PBX is okay, nothing fancy, you could build your own Asterisk / FreePBX / 3CX hardware. We had quotes of around £30,000 from various suppliers for a hybrid system that would re-use some of our old phone system's wiring, but in the end just chucking the whole lot out and buying the PBX and 60-odd new Yealink handsets was under £10,000. We did a (separtly budgeted) network upgrade at the same time, so we have nice new PoE-capable switches for powering the handsets and two ethernet points in every room, but even if you only have one network point per room and have to find a mains socket to plug the phone handsets in you should be fine. Do just bear in mind that the handsets tend to be 100 mbps pass-through links rather than gigabit - you either have to pay a bit more for gigabit-capable handsets or accept 100 mbps speeds to desktops that use pass-through connections from your phones.

I have an event to film this Friday that I was thinking to get a 360 degree camera for - I've been wanting one for a while, but the Bublcam I was going to order sold out the instant it was available. The models currently available from Jessops would seem to be the 360Fly, the Kodak SP360 4K and the Ricoh Theta S. Both the 360Fly and the Kodak have a field-of-view of around 240 degrees. The Kodak has a neat mount available which houses two cameras, but that then requires stiching two lots of video together (it also requires buying two cameras). The Kodak's included stiching software doesn't seem to be all that great, and third-party stiching software is around the £500 mark, about the same as the camera itself. I think I'm going to wait until something that does in-camera stiching becomes available - the Bublcam or the Giroptic. From what I can make out, once you have your footage in an MP4 file as a equirectangular projection it's actually quite easy to do whatever you want with it. YouTube will host it for you, or you can host it yourself and write your own 360 degree player - actually way easier than it sounds as there's a Javascript library function to do the projection part of things for you, you just add the user interface controls.

I don't know enough about music / sound to be able to tell the difference between a high-end and low-end microphone, but to me our 4-way set of Chord mics sounds just fine. They now do a 6-way version: http://www.amazon.co.uk/Chord-HU6-H-Wireless-Microphone-System/dp/B00NRNYRKQ/ref=pd_sim_sbs_267_4?ie=UTF8&dpID=41t81ZidwXL&dpSrc=sims&preST=_AC_UL160_SR160%2C160_&refRID=1P23PDH04H0W65HABC74 We've found our 4-way unit to be robust and reliable over the past couple of years - we've mounted the mic mixer in a stage box, complete with power supply and a drawer for the mics, so we can just pick the whole thing up and carry it to wherever it is needed. As a complete all-in-one solution (you don't need a mixer, you can use 6 separate outputs (one per mic) or one mixed output, with volume controls on the front of the rack unit) for under £400 it's really not bad. If nothing else, you could get that unit for general day-to-day use and save some wear-and-tear on whatever snazzy high-fidelity kit you get.

We network boot Windows install images via PXE with Serva, then automate the windows install process with an unattend.xml file. At the end of that unattend.xml file we hand over to a VBS script that installs all of our software.

FreeNAS is based on ZFS. I imagine it should be quite possible to take apart a ZFS volume and put it back together again on different hardware, but you might want to test that before you go relying on it.

If you use a NAS that uses Linux's software RAID (MD RAID) you should be able to move the disks around between different hardware with no problem.

I saw the Padcaster at BETT this year - I haven't managed to find an excuse to buy one yet, but I'm working on it. They seem like excellent bits of kit, letting pupils do the whole storyboard / film / edit cycle pretty much on one device.

Something I want to try now the Raspberry Pi does wireless and bluetooth: hook the Pi up to a large screen, run a game. Have it act as a wireless access point and/or bluetooth device that people can use their phones to connect to. Have the only thing the access point serves be a single-page HTML game controller widget - user comes along, connects to wifi access point, game controller appears, plays game.

My guess is that this is likely to cause more issues than simply leaving the drives in place while you re-format the boot drive. If you remove both drives and reboot the machine the RAID controller might decide the array has disapeared and do away with any settings. I've never had a problem re-installing an OS on a boot drive while leaving a data drive intact, although obviously make sure you double / triple check which drive you are reformatting before you actually hit "go"...

TTS do a non-bright-yellow "pro" version: https://www.primaryict.co.uk/easi-speak+and+easi-speak+pro_c72_p1.htm I've used these in several diffrent classes / schools now and they really are very useful, versatile devices, although as pointed out not really up to the quality you need for voice recognition (not quite sure why, the actual recording generally sounds clear enough - sample rate?). If you're holding them in your hand they do tend to pick up every little rustle and movement you make, but you get the hang of using them after a couple of quick tries. The fact they plug straight into a PC as a USB storage device is rather handy, too.

Hacker News' discussion raises some valid points: https://news.ycombinator.com/item?id=11188445 When Google say "SSD drive" they don't mean something they bought from Amazon, they buy memory chips direct from the factory and make their own drives with their own firmware. These aren't drives inside laptops and similar, these are in datacentres, typically running at around 35 degrees celcius. I also doubt most of those drives are in a RAID array, Google designes its own high-availability storage architecture, and has moved on rather from RAID arrays. Reading that article and various comments, I think we're doing okay with our array of "consumer" Sandisk Ultra II SSDs currently in one of our servers. This is in an HP server with an HP Smart Array RAID card, updated to the latest firmware version. In particular, that firmware update has features to let the RAID array use SSDs properly. Regarding the article above, the comment about over-provisioning seems relevant. Over-provisioning in case of data loss doesn't seem to be worth the bother (SSD drives seem to fail by age rather than by usage), but having over-provisioning in place to avoid write slowdown when the drive itself does garbage collection might be handy, depending on application. Our RAID controller seems to let you specify over-provisioning controlled by the RAID controller, so you don't have to worry about the feature on the drives.

We currently use XenServer, so moving to something based on Azure is going to need a move to Hyper-V. We plan to move over when the new Server 2016 with all the snazzy storage features comes out.

From what I can make out, it's actually a good database from the point of view of rapid application development, letting non-programers put together a database with the features they actually want. However, scalability seems to be somewhat limited - you can add more web presentation servers, but they still all have to utilise a central main database server.

Just to add: the larger screens (65" and up) are quite heavy - check the load-bearing capacity of your wall before you decide what mount you'll be using, you might need to go for something that bears the weight on the floor.

This might depend rather on how your school operates, but we find we wind up re-locating equipment a fair bit, hence we're tending to buy LCD screens as they can be placed in wheeled stands and simply moved to where you want them. We recently bought a Promethean 55" screen with a motorised stand (so we can adjust the height, for different ages / height of children) and built-in OPS PC (8GB of RAM, SSD, Core i5) for £3,600 (price includes VAT). No installation charge as there's no installation, we just have to slot the screen on to the stand and away we go. The OPS PC hasn't turned up yet so we can't give a definative this-is-the-setup-to-go-for yet, but certainly the Prometean boards currently seem to be the best value LCD screens out there at the moment - not the absolute cheapest, but the cheapest good-quality, proper 10-point multitouch and with OPS slot support.

We would want to be able to run a bunch of virtual machines on local physical servers and have some kind of lovely simple just-enable-it-and-it-works system take care of mirroring those VMs to an offsite location, hopefully in as near to real-time as possible. As soon as a VM (or a whole physical server) breaks, the remote service notices and spins up a running VM. Users never notice. Also, the ability to snapshot VMs and restore from previous snapshots would be good, as well as the same for file systems - have a versioned backup system that lets users get to previous versions of their own files as they want. We currently have around 3TB of VMs (OS images) and around 6TB of user files, what sort of price would you be thinking to charge for that kind of size of system?

Me? No, we have a custom component that uses OAuth to authenticate with the VLE - users log in to the VLE, which takes care of permissions. If this is on a machine inside your firewall you can simply make it a domain member and assign appropriate file permissions to the staff / SLT / etc folders, then have the web developer make sure his applications check the current user permissions before giving them access to file / applications in a particular folder. If this is on a Linux server you can make that a domain member to via Samba.

However your in-house developer decides - as pointed out above, the most common solution would be to have the website check group membership for permissions to various sections / functions on the website. You think / hope there'd be a nice framework available that the developer is using for this rather than simply writing it from scratch - even if they simply used file system permissions on the folder structure on your web server, that would be quite simple to implement.

Just to add another system / supplier (the one that we use) - CJM's School Manager: CJM Software

Any luck? If doing "Save As..." works and if you can't figure out which file it's looking for as suggested above, your next best bet is to automate the saving-as process for as many files as you need to - I would suggest an Auto-It script for this, left running overnight to work its way through a folder's worth of files.

I've seen a situation before where documents were created from a template, then when that template was no longer available the document took ages to open while Word tried to open the template and waited for 2 minutes or so for the template to open. If the template itself is still there then maybe checking file permissions would be helpful (i.e. you can see the template, but can the users in question), and I also found that, once opened, re-saving the document with "Save as..." got rid of the hard-coded reference to the template, so the document opened just fine after that.