* Posts by Silas S. Brown

13 publicly visible posts • joined 17 Aug 2013

UK's National Museum of Computing asks tunesmiths to recreate bleeps, bloops, and parps of retro game music

Silas S. Brown

I have a MIDI to BBC Micro converter

which can generate BBC BASIC code from any MIDI file you'd care to create, fitting up to 3 notes per channel via envelope arpeggiation, although I didn't code any support for the noise channel (but happy to accept patches).

Alternatively RISC OS has 8-channel polyphony (and runs natively on the Pi), but viewpoints may vary as to whether that can be counted as "retro".

Copper broadband phaseout will leave UK customers with higher bills and less choice, says comparison site

Silas S. Brown

Low-income households could lose out

The trouble with those growing percentages of people opting for fibre is they may leave out low-income households for whom price is a much more important factor. If fibre is too expensive, they may have to go mobile-only and be subjected to data caps (whereas their previous ADSL contract gave them unlimited data). Has anyone in authority thought of giving fibre discounts to people on benefits? (although not everyone who needs them knows how to claim them, but it's a start)

You've got to be shipping me: KatherineRyan.co.uk suggests the comedian has diversified into freight forwarding

Silas S. Brown

Re: Can somebody explain the economics?

I can understand why they'd do that if it cost them £10, but I still don't understand how anyone thinks it's worth $2,060 to do that. Unless they plan to do something else later, but homenet.org is still redirecting to Hotels in Vienna after more than 3 years. Seems like the catcher companies have a good business from highly-speculative investors though.

Silas S. Brown

Can somebody explain the economics?

I find this hard to understand: why would someone pay good money to buy an expiring domain and then redirect it to a completely irrelevant business? Yes there might be a small number of new customers from those "oh look where my old link or bookmark takes me now" moments, but are we really talking about enough profit to justify the cost of doing this? And if what they really want is to sell the domain to a higher bidder later, then they don't make it at all obvious it's for sale. Or could the redirection just be a front for some more clandestine operation that somehow requires a long-established domain?

The same thing happened to homenet.org in 2017. That one used to be a hobbyist dynamic DNS site for home servers (with 105 users), but when it expired, DropCatch sold the name to a certain resident of Kiev who somehow thought it was worth bidding 2,060 US dollars just to redirect it to a list of hotels in Vienna. Apparently that buyer paid similar amounts to win other domains and redirect them to random other hotel sites. And yet I don't think any of the old Homenet users are going to say "wow, I should book a trip to Vienna" instead of "bother, the server I wanted has moved". How can it possibly be worth paying thousands of dollars per domain to redirect that kind of traffic?

(I suppose buyers might think there is a small possibility of gaining a little extra Google PageRank from other sites that fail to update old links to the domain they bought, but is this negligible PageRank really worth over 2000 dollars? How many non-updating incoming links did they think HomeNet had?)

Heads up: From 2022, all new top-end Arm Cortex-A CPU cores for phones, slabtops will be 64-bit-only, snub 32-bit

Silas S. Brown

Re: Presumably also less power usage ...

Hopefully by means of an emulator. Less efficient than hardware, true, but if few users need to run historical applications then I for one would be happy with an emulator instead of taking up everyone's chip space whether they use it or not. As long as there will be such an emulator.

ICANN begs Europe: Please fill in the blanks on this half-assed GDPR-compliant Whois we came up with

Silas S. Brown

Whois lookups used to be useful

If I'm thinking of dealing with a company, charity or other organisation, I used to rely on WHOIS to let me check that the domain really does belong to them. Obviously I'm not after anybody's home address to snoop or whatever, I'm just after a bit of reassurance that a third party (the registrar) was convinced this website owner is the "real deal". Office address would be fine by me (if they have an office). And creation / transfer dates give you some idea how long they've been around (at least on the 'net): "a few years" is better than "since last week". Not infallible I know, but every clue helps when you're trying to figure out how much risk to take.

Nowadays, even the genuine domains of some large organisations will WHOIS-lookup to a "privacy service", which was probably a mistake (did somebody forget to tick the "no we don't want this" box?) - they could easily have put their headquarters address in there, to help establish the domain as genuine, without disclosing anything they weren't already making public on their About Us page. What a missed opportunity.

Go Huawei, Android: Chinese telco biz claims it will spread Harmony OS for smartphone to devs come December

Silas S. Brown

Symbian and Windows Mobile weren't dismal

If anyone's going to call the performance of Symbian or Windows Mobile "dismal" then they ought to clarify that this refers to recent years i.e. after iOS and Android came onto the scene. Back before the rise of iOS and Android, let's not forget Windows Mobile had a 25% market share and Symbian's approached 50%, not dismal at all.

Sueball locked, loaded and pointed at LinkedIn over iOS privacy naughtiness

Silas S. Brown

Free/Libre and Open Source advantage

I made an Android app that reads the clipboard on startup without asking. It does it to see if the clipboard contents is a URL, in which case it pre-populates a text field with that URL. I could have added an extra Paste button, but then I'd have to worry about issues like "would the button take up too much room on small displays" (or if it's hidden, would users be able to find it), and "would users be confused if pressing the button causes an error message because the clipboard contents is not something we can handle—would it be better to simply not offer the option in that case". So I decided (rightly or wrongly) that reading the clipboard without asking would be the least-bad thing to do by default.

But my app is Free/Libre and Open Source Software. If you don't trust me, you can read the source code (or trust that others will have read it and called me out already if necessary). To make really sure, you can compile it yourself and verify that the bytecode you get is the same as the bytecode on the Play Store version (if you use the exact same compiler version and options as I did).

Unfortunately for LinkedIn, they have so far not been brave enough to show the source code of their app, so it's not as easy for them to use the "just read the code" defence. (Perhaps they'll now invite lawyers to look at it behind closed doors, but publishing it could have saved that trouble.) LinkedIn has open-sourced some minor components, but not the app itself. I don't know the business reasons for that decision, but it seems clear to have this PR disadvantage at least.

That said, the iOS ecosystem itself isn't exactly helping. It doesn't make it at all easy to verify that what you get from the App Store is the same as what you can compile from source code, plus its incompatibility with certain licenses (along with the general high financial cost of being an Apple developer) seems to have resulted in an environment somewhat devoid of Free/Libre and Open Source Software (so iOS users are missing out on some of the best software). But in LinkedIn's case I have not been able to find their published source code for the Android app either, sorry to say.

Publishers sue to shut down books-for-all Internet Archive for 'willful digital piracy on an industrial scale'

Silas S. Brown

Shutting down the Wayback Machine too?

The sentence "The publishers hope to shut down the dot-org, we understand" implies that they hope to shut down the whole of archive.org, not just the eBook library. The whole of archive.org includes a lot of resources that are nothing to do with that eBook library, most notably the Wayback Machine for showing old versions of websites. The Wayback Machine has been useful in legal cases requiring third-party evidence of what some company's or government department's website said before they changed it, so I would imagine a lot of lawyers would not like to see it disappear. It has also been useful for rescuing "dead links" on Wikipedia citations, and I'd expect some (possibly valuable) Wikipedia articles would end up being deleted if it's no longer possible to show (from a neutral and trusted third-party) what sources used to say.

Is The Register allowed to say on what basis they understand that the publishers hope to shut down the entirety of archive.org and not just the eBook library?

China and Taiwan aren't great friends. Zoom sends chats through China. So Taiwan has banned Zoom

Silas S. Brown

Use Jitsi instead

Who knows what happens in Google and Microsoft's servers, but Jitsi and similar Free Software alternatives mean you can self-host.

IT now 10 percent of world's electricity consumption, report finds

Silas S. Brown

Raspberry Pi to the rescue?

My ARM-based Raspberry Pi home server takes less power than the router and it's still able to run my Web Adjuster system to annotate Chinese websites. Why isn't there a company doing Raspberry Pi co-location? Yes I had to spend more time thinking how to set it up using lightweight tools, and I suppose "time is money" in a corporate setting, but on the other hand I did end up with something more scalable as a result (if it can serve my userbase on the Pi then how many could it serve on a bigger machine), so perhaps it would be a useful exercise to develop for Raspberry Pi servers first?