VogonPoetry

"population count" is often associated with hamming distance with uses in information theory, coding theory, and cryptography. There is another use, which I think is what it was really used for - cosine similarity.

The usual definition for two vectors A and B is A.B = |A||B|cos ø. If A and B are binary it becomes, A and B = popcnt(A)popcnt(B)cos ø. The similarity measure is how close cos ø is to 1.

Using modern vocabulary, with a message and a set of search criteria using a word embedding on the message, calculate the similarity to each search using, cos ø = (A and B)/(popcnt A * popcnt B). This is only be vectorizable when there is an intrinsic popcnt instruction. Results that have > 0 values for cos ø mean there is some similarity.

So, to capture, (re)scan and flag all messages that contain certain keywords, design a system to generate a KEYSCORE. When eXtended to reduce false positives, build a follow on system called XKEYSCORE.

The last paragraph is pure speculation, but I think it has merit. Worth a try on your own data?

Okay, I'll bite. Also see my other comment.

You said "mouse", is it a gaming mouse? Is it configured to issue a bazillion clicks if you press and hold "left" button - for a stupidly small number of ms, taking no consideration of you if you are playing a game? Does it also send "wiggler" (WASD) keyboard presses at the same time? Either of these will ruin any chance of you doing "undo", unless you also "undo" a bazillion times. This behavior will likely make a lot of other things behave in quite weird ways and be "slow" to respond. If you've got a gaming mouse doing auto-clicks, auto-kbd strokes for non-game apps, then the UI designed for humans is going to behave weirdly.

FWIW, the placeholder text can never really disappear, if you remove it all, you get the placeholder "Text" back again. You can replace it with a single space, but that also has a hit zone of more than 1px both H & V. Maybe you have found a bug. I do think there are issues with font background / foreground colour selection that might make it "invisible", but the hit zone will still be there.

I use an ancient mouse, "Microsoft Comfort Optical Mouse 1000" and an Apple Trackpad. I cannot reproduce what you describe.

What mouse are you using? What additional HID (Mouse, KBD) extensions do you have installed?

If the PDF doesn't contain form completion elements you have to use the "painting / diagramming mode" to add annotations. However, I often find the "form enabled" PDFs to have horribly mis-aligned boxes with odd choices for fonts and inconsistent behavior (check marks or choices for example).

I do agree that the annotation feature isn't like Omnigraffle or draw.io diagramming apps that are based around a mouse; the interface does seem more suitable for a touch (pad, screen). You choose an object to add, then get to move it into place rather than selecting an object to add and then use the mouse to place and size it. The annotation feature also works in a very similar way on an iPad / iPhone.

If you tap (click) and /hold/ anywhere in the hit zone of an object the cursor will move to a hand and you can move it anywhere. For a text box, the hit zone it is almost anywhere on the text - yup, it is more awkward if you delete all the text! The resize / reshape hit zones for objects are centered around the "dots" when the object is active. A single tap (click) will edit text.

I am not sure what is happening when things disappear and cmd-Z doesn't undo, I've not experienced that.

This markup feature is also available when annotating image attachments in email.

It would be nice to be able to group more than a single set of objects and keep them grouped.

I've used the annotation feature quite a lot to highlight errors in web pages and give sequence instructions on how to do something on a poorly designed site.

Didn't have the same feel about the piece. But based on the parent comment, decided to also test it, using Pangram free "credits". For me it reported that only sections of it (30%) have been "touched" by an LLM. There is 60% of human written text.

Pangram does not report this as 100% LLM generated. Why do I use the word "touched" - because there are some specific details that must have come from some existing text and are not elsewhere.

How did I do the check? Please do check my results! I copied just the authored prose from the page, carefully removing any comments and site text that got mixed in. I did not have it analyze the "URL".

In looking at the details, a large part of the story is flagged as all human generated.

I've personally used LLMs when I've been very upset. I used it to rewrite / edit my rant into something with a more balanced tone. The modified text got what I wanted, my original rant would almost certainly not have.

So I do think this was a thought out piece that was largely written by a human who used an LLM as an "editor" to polish the writing. Someone took time to think about this and write it.

Bloom Energy makes fuel cells that can run on natural gas. This site would need a very large number of them to replace gas generators. Each "bloom box" can create 325kW and are about 1m x 1m x 2m in size.

They make sort of whine/squeal noise when operating. Not sure of the dB level.

San Jose Ice Rink has had a bunch of these installed out back for about 10 years I think.

Another comment mentions a KryoFlux. A cheaper alternative is a GreaseWeasel (£30, $30). The 4.1 version even has USB-C as a connector. You will need an external power brick with a 4pin Molex connector for 5.25 drives. You can buy these standalone or they might come with a SATA/PATA/IDE Hard Drive to USB converter. Can also use an old PC power supply if you fake the power on signal.

I've had quite a lot of success, almost 100%, over quite a long time - both 5.25 and 3.5. I did use older drives intended for "desktop" machines. The drives from Japanese companies seem to be the most reliable; NEC, TEAC. The 3.5 USB Portable diskette drives are based around laptop floppy designs. You won't be able to find a direct 5.25 to USB adapter (or modify one to work) as the speeds / data rates are all wrong, so you will need something like a GreaseWeasel or old PC for those.

I've used different reading mechanisms over the years; an old PC running Linux with a real floppy chipset + custom raw reading code, custom hardware with an F82077 + Atmel + SDCard, and most recently a GreaseWeasel.

The floppies I've recovered were stored relatively well, but were mostly my and family members data - not commercial floppies. If there is any copy protection on the floppies you are trying to read you may need to capture the flux transitions. For these, I'd check archive.org to see if someone has already generated a copy.

For 5.25 floppies fingerprints and surface blemishes can be softened / removed with ISO Propyl and a Q-tip. Floppies designed for cleaning the heads of drives seem to now be very expensive / non-existent. Manually cleaning the heads of a 3.5 drive is way more difficult. I've got ideas about using Face mask / filter material to create a cleaning tool, but it is on the experiments to try list.

Even back when floppies were a thing, I found that 3.5 drives were more un-reliable between machines than 5.25 drives. I think there was more variation in head alignment and track positioning. When I've had issues, I've always been lucky in that switching the drive vendor / model made reading the floppy possible even if it took multiple attempts or was very slow.

Good luck.

Your site does not currently seem to be reachable / responsive when I try to reach it from a US Comcast IPv4 address - you are not advertising IPv6.

Edit / Update: It was Apple's Private browsing mode that causes it not to respond. I can now see it when this is disabled.

I wonder if "Piano Mover" is a reference to the Goon Show episode called "Napoleon’s Piano" [1]

If you like absurdly silly comedy; Harry Secombe, Peter Sellers and Spike Milligan, you should track these down. These were made in the 1950's for BBC Radio.

[1] https://goonshow.org/goonshow/s6p4/

There are a lot of articles written about early machines available from the (British) Computer Conservation Society. The CCS journal is available online at [1]. One article (with a link to a GitHub) discusses making an FPGA version of Colossus.

There are many articles written by people who were involved in early computing in the UK.

[1] <https://www.computerconservationsociety.org/resurrection.htm>

It is still possible to adjust the speed in QuickTime Player: from .5 to 2x. Not available via Quicklook, but definitely there in the player, either in the menu bar or by clicking the >> in the video window.

This is fantastic info, thank you. I've now set mine to 5.

The post was fixed about 30 seconds after making it - due to the *s being interpreted as italics. It is a shame there isn't a preview button when composing posts.

Does this also mean only using "standard" parts? Or does the manufacturer have to over-produce the parts for, lets say 7 years, and then warehouse and ship those parts, probably multiple times. Or keep a low rate production line running for 7 years? What happens to the parts that don't get used? Are they scrapped?

That "what if" cost is going to be built into the cost of the laptop. Repairability doesn't always keep the cost low. The purchaser will definitely have to foot the cost otherwise it isn't sustainable.

That 2mm uses at least (2*335 + 2*235) * 2mm * 1mm = 2,280 mm^3 more material for the case. (a wall thickness of 1mm)

Please can you explain the evidence that this is generated content.

Yes, the site is new, but other posted articles are 100% consistent with the author wanting a guaranteed level of local inference with large models.

What I read was written by a skeptic who took claims and systematically addressed a number of issues. The debunking was concise and used simple sentences.

The boxes in the pictures were, to my eyes, generated manually using the macOS Preview annotation feature. They are not well aligned. I've used this technique many times to general overlays. If this were me, I'd get called out. I like nicely spaced and proportioned boxes! NB: iFixit tear downs are mis-aligned as well and it bugs me.

People have a distasteful habit of assigning others into boxes. Particularly if that box currently has a negative connotation. Boxing is a primary tool of the; misguided, bullies, sycophants, censors and those with an unspoken agenda. Humor: Which box applies?

I took the "granularity doesn't cut it" comment to mean there aren't enough entitlements to eliminate the need for custom SBPL. Followed by a sentence about apps that have temporary exception SBPL. Combining the two seems to imply that if there were more entitlements the custom SBPL might not be necessary. In the followup you noted; the split in reasoning and evaluation is rough and potentially not needed. I read this as a conclusion of wanting to do something, but could not as there were not enough entitlements to make it work, so custom SBPL would be necessary.

I am not sure using sandbox-exec is a good security architecture for AI agents. It sure is convenient and available to everyone right now. I've made another comment elsewhere in this discussion about what I think "deprecated" means - it is a sharp tool that could break if not tracking everything that changes, including every change in a SW update. It is also easy to get wrong if there is not a "(default deny)" in the profile. An agent could escape if they can find a mach service or some other system call coordinated proxy service. Java, Silverlight and Flash had backdoor communication mechanisms with other instances of themselves that could be abused.

Does this mean you tried to ship an App in the Apple App Store but could not because of some restriction?

The Sandboxing and Entitlements mechanisms are very different. Sandboxing can only drop access to resources, it cannot grant access that was not already there [1]. Entitlements are all about giving additional selective privileges or to make the sandbox NOT remove access (like full disk access or debug ability ). Entitlements are bound to processes only and are non-transferable. This is in contrast to a capability based system where they can be passed around. Reasoning about capabilities is challenging because analysis effectively requires global knowledge of the system. Binding entitlements to libraries or Frameworks would turn them into capabilities.

[1] a GUI app can restore access to files by using a trusted external selection process.

Edit: change footnote reference to prevent markup error.

The runtime engine is not known to be turning complete. It has no expressions and cannot loop, only forward jumps are permitted.

"sandbox-exec" is deprecated in the sense of "please don't use this method to run sandboxes" rather than the mechanism going away.

If you are using "sandbox-exec" then you are likely maintaining your own seatbelt profile. Keeping those up to date can be challenging, especially for 3rd parties as any changes to underlying Frameworks and libraries can break a hand crafted profile.

If you are using it to secure your own stuff and accept this and not complain, even for minor SW updates, then you are going to be fine. Don't ship things to 3rd parties without also accepting this. That is what this deprecated means.

I have received feedback from Vox. The article has been updated with a new leading paragraph indicating the fictional nature of the article.

I have received feedback from Vox. The article has been updated with a new leading paragraph indicating the fictional nature of the article.

I've written to <voxmeditantis@gmail.com> about how deceptive it was to put the Editorial Note at the end instead of upfront. I stopped reading because sections felt fabricated - but it was presented as an oral history or actual interview. What a terrible way to present the work of a pioneer.

An Editorial Note is at the bottom (as others have now noted), it should have been at the top. Had I not seen other comments I would likely have believed everything was made up. This is a terrible way to recount the memory of Frances Allen.

There is something off about this piece. Particularly the section that starts "You passed away on your eighty-eighth birthday – 4th August 2020. Do you reflect on mortality?" I stopped reading after that.

Now that I've been recalling more memories of this, I do remember there being encoding or "escaped" character issues - particularly with brackets and parentheses.

There was another device between the BBC Micro and the "Versa Braille" unit. The interposing unit was a matrix switch that could multiplex between different serial devices - I now suspect it might also have been doing some character escaping / translation.

For those not familiar with Braille, it uses a 2x3 array (6 bits) to encode everything. The "standard" (ahem, by country) Braille encodings are super-sub-optimal for pretty much any programming language or mathematics.

After a bit of (me)memory refresh, in "standard" Braille you only get ( and ) - and they both encode to the same 2x3 pattern! So in Braille ()() and (()) would "read" as the same thing.

I now understand why you were asking about the software used. I do not recall how we completely worked this out. We had to have added some sort of convention for scoping.

I now also remember that the Braille terminal aggressively compressed whitespace. My friend liked to use (physical) touch to build a picture, but it was not easy to send spatial / line-by-line information to the Braille terminal.

Being able to rely on spatial information has always stuck with me. It is for this reason I've always had a bias against Python, it is one of the few languages that depends on precise whitespace for statement syntax / scope.

Yes, mostly raw TeX, just plain ascii - not specially coded for Braille. This was quite a long time ago, mid 1980's, so not long after TeX had started to spread in computer science and maths communities. My friend was using a "Versa Braille" terminal hooked via a serial port to a BBC Micro running a terminal program that I'd written. I cannot completely remember how we came to an understanding of the syntax to use. We did shorten some items because the Versa Braille only had 20 chars per "line".

He is still active and online and has a contact page see https://www.foneware.net. I have been a poor correspondent with him - he will not know my HN username. I will try to reach out to him.

I did a maths undergrad degree and the way my blind, mostly deaf friend and I communicated was using a stylized version of TeX markup. I typed on a terminal and he read / wrote on his braille terminal. It worked really well.

After writing this I decided to look at my shortcut. The action seems to have been a simple "get directions to <place>" and sent verbatim to Siri.

I was not able to edit / update it! However, there was now a new "maps" option for `Open <type> directions from <Start> to <Destination>`

Where type can now be {driving,walking,biking,transit} and <start> is Current Location by default.

After updating, this now seems to correctly set actual driving directions, even if I'd previously set up a walking route!