I don't want to downvote this because it's interesting, but the tone "you're an idiot if you don't already believe my extremely niche view" works against you.
They barely mentioned your website (fourth in five urls, mainly talking about indieblog.com and kagi.com/smallweb), so "That is my website!" is confusing and makes it seem like you're autoresponding to a keyword.
Yeah, it didn't confuse sensible people who are capable of putting themselves in someone else's shoes.
I did a Small Web search at Marginalia and was immediately pointed to sites that claim that I and everyone in my political party are literally the spawn of Satan--I really don't think it's my thing.
I helped develop the ARPANET back in 1969-1970 while working for the UCLA Comp Sci dept, got a brief mention in RFC 57, hold several network patents, and was on usenet before the usenix conference where we voted to call it that ... I'm bemused by all the people who claim that boomers are technologically inept (I think they have us mixed up with our parents). Anyway it's been a heck of a wild ride and didn't end up quite how JCR Licklider envisioned it.
I wrote this in response to one of the reviews, so I'll share it with you since you asked. :-)
I worked on an ARPA-funded speech understanding project in the 1970's at SDC--it was definitely driven by military interests. One time some of us techies were in our soundproof lab drinking wine and eating cheese and crackers when our manager brought to the big picture window an Iranian general bristling with medals--they both looked extremely unhappy.
I also worked on ARPANET development at UCLA from 1969-1971, and there was none of that. The driving motivation was ARPA-funded researchers at universities being able to readily share their work. Is ARPA funding researchers at universities an issue that can be written about? Of course, but it has nothing to do with the ARPANET per se and isn't part of the story that this book is about.
Oops--I left out a critical part of the SDC story--we were in the lab because it had an incredible sound system featuring a pair of high end AR-3 speakers. I don't recall what we were playing but I'm sure it sounded wonderful.
We also did real work in that lab of course ... mostly recording things like "What is the surfaced displacement of the Lafayette?", which our primitive system running on (pathetically slow by today's standards) Raytheon 704 and PDP-11 computers would attempt to parse and answer. The text of course was selected for the sake of obtaining a grant from the USN.
This early work, funded by the military, laid the basis for today's ubiquitous speech understanding systems. Are there issues with fundamental research being funded by the military or, say, big pharma, rather than as part of a direct planned effort by society to achieve social goals? Sure, and much can and should be written about that, but it's not the subject matter of this book.
If it could actually parse your speech and come up with an answer in 1970, it must have felt amazingly futuristic. Star Trek was only a few year old at that point. Thanks for sharing that.
The parsing broke speech into phonemes--actually a string of candidate phonemes, each candidate having an assigned probability. It made a lot of mistakes--it couldn't generally distinguish between "a" and "the" in rapid speech, and the semantic phase didn't help disambiguate those. It worked better for female voices because they have an extra formant. It didn't work well if the speaker was intoxicated--we learned this from some anomalous results that one researcher dug into and discovered that there was a "knee" in the data--it turned out that our late night speaker Bill (a giant bearded guy who wore overalls that he ordered specially from Iowa IIRC and was known as wabblezabble) had taken a break, during which he drank a considerable amount of beer, on the hypothesis/excuse that it would make his speech more, er, fluid. It had the opposite effect--the automated recognition was consistently better before the break than after.
Coming up with the answer required doing a nondeterministic parallel search of the candidate phonemes through a DAG of phrases--the problem was contained because the DAG was highly restricted to the subject matter, in this case facts about Navy ships. This was a pilot and the dream was to have a much more massive semantic net of the English language. We had linguists and a resident lexicographist (he distinguished this from a lexicographer, though the dictionary says they are synonyms--but lexicographists know better than dictionaries created by lexicographers, heh heh) working with us. The parsing code that dealt with the audio signal was written in FORTRAN and assembler, IIRC, but all the language stuff was written in a local version of LISP. Jeff Barnett, on our team, was the author of SDC's LISP2, but I'm not sure that's what we were using. He was working on developing a more performant algolish LISP called CRISP when I left. Jeff had written the parallel search algorithm, which had a "knob", as he called it, which was a floating point value that controlled the depth first/breadth first balance--any possible balance could be achieved by dialing the "knob". This was needed because it took too long to do an exhaustive search--it bailed with an answer as soon as it found one that passed some threshold. Anyway, it required recording onto tape, digitizing it and feeding it to the minicomputer, running many passes, feeding the results into the LISP program running on a mainframe, waiting an indeterminate time to make a match against a highly restricted vocabulary--more of a grind than futuristic. I remember when programs like Dragon Speech showed up ... way advanced over what we had, but still needing to be trained on a specific speaker. Now we have realtime language translation in our pockets. The other day I accidentally turned it on and my friend at the other end of the line asked who was speaking Spanish ... everything I said was being repeated in Spanish.
BTW, when I left SDC because I wanted a break from work, they offered me a spot with their new development called EFTS, but I was pretty set on leaving. EFTS--Electronic Funds Transfer System--is the backbone of all of today's digital money transfers ... ATMs, ACH, etc. I really missed the boat on that one.
P.S. In trying to remember why Bill (aka Billy) also had the nickname wabblezabble, I managed to remember his last name, which yielded his initials WAB (at UCLA initials were used as login names). I found this lovely obit which very much fits the guy I knew: https://www.legacy.com/us/obituaries/latimes/name/william-br...
Auto-responding when Google Alerts spots your keyword on a new site would help promote your website.
I missed that "That" meant "the site that matches my username" so I had gone through a process of "how is this guy claiming to own https://kagi.com/smallweb" and "did he mean indieblog.page/random" and "why is he also saying he wrote something at unsungNovelty.org which doesn't match either of those". I never did figure it out, I opened all the links to check the author and finally got it. Sorry I was dumb. But antecedents are important.
No worries. But If I am behind numbers, my website will be completely different with post cadence far more regular. Not saying regular posts are bad. In fact, I am currently working on finishing my writing backlogs. But still, not everyone is behind numbers. Especially like SEO/google kind of numbers.
So frustrating when every conversation leads to R vs D. Doubly so in this situation since both bills that got us to where we are today had overwhelming BIPARTISAN support and were signed into law by presidents Clinton and Obama…
If individual party members voted against the party line more often there would be less of this kind of discourse. But the reality is that we have a deeply entrenched deeply divided two-party system. There are very few politicians who don't toe one line or the other and endure. But in this case it's a core tenet of the republican party platform to eliminate the administrative state, including strategic investment and reserves.
That's not much of a source -- a 100-karma user in 2020 based on "I've known this for a long time. A quick google confirms that many people think the same." I don't believe it is true.
I mean... Maybe the things I'd LIKE to work on are getting my car around the race track faster. Very few people will pay me for that - especially if I'm not a very good driver. But I enjoy it immensely. I'd MUCH rather do that than work.
And right now, due to having to work, maintenance on my house is a bit behind.. Would also prefer to catch up on that - but again, no one is paying me to do that.
> if the rate of fraudulent labs is even as high as 10% you should expect that any viewpoint that it’s widespread would be drowned out by views that it’s not real.
If "it" is fraud here I would expect the viewpoint that it's widespread to be less and less drowned out as it approached 10% since everyone would know that it's real. I think I'm misunderstanding the sentence.
No, the guys at fraudulent labs and the guys at honest labs will both claim no fraud. The only ones who will claim fraud are those who cross over. So you’ll get a vast majority telling you it’s not happening and a tiny minority (even when as high as 10% are fraud) telling you the fact. All rare things have this effect. There will be so many people telling you it’s not real “as someone in the field”. They will be adamant about it. You need someone who has seen both.
To be clear, not “as it approaches 10%”. I mean “even as high as 10%”.
Not I didn't say that it isn't real, I just said it's not that rampant, and I was referring to the egregious examples in the article and the anecdote. But it's a sliding scale and being selective in presenting research results is a gray area that requires us to think about reproducibility and make sure we have good processes for discouraging, detecting, and correcting those kinds of dishonesty, too.
In my field (programming languages and compilers), conferences have started to adopt artifact submissions which include source and build instructions that are then independently replicated by volunteers who serve on artifact evaluation committees. It's a good step in the right direction.
And I'll also caution, again, that reasoning based on anecdotes can be distorted both ways. A tiny minority of people can claim the sky is falling and make it sound like Science is threatened by crisis. Those voices get amplified by the media and the attention economy and get blown out of proportion. What Science and institutions need is mechanisms that are fair and well-resourced and taken seriously. Those mechanisms and the people who volunteer time and effort to keep them running are the real heroes of Science, and breathless articles and anonymous anecdotes on the internet do not make up an accountability system.
If you see fraud, then REPORT IT to someone who can actually hold them accountable. Unsubstantiated and undirected accusations just spread distrust and do not increase accountability.
>And I'll also caution, again, that reasoning based on anecdotes can be distorted both ways...
I will go further and say that if you can clearly see how the incentives are aligned, even anecdotes are not necessary.
>If you see fraud, then REPORT IT to someone who can actually hold them accountable. Unsubstantiated and undirected accusations just spread distrust and do not increase accountability.
Mmmm..are you..for real? I mentioned incentives just above, so let us look at it. What is the incentive of this "person who can hold them accountable"? Do they really want to cut down the "output" of the institution by insisting on 100% honest? Do they really want to lose to competition which might allow such practices?
I think not. I think if this is reported, chances that the report will sit in some shelves or end up in some dust bin doing nothing. And the person who report it would suffer one way or the other.
I was single for 10 years after college. Then married for 7 (very very badly). Then single for another 10. I thought I will spend the rest of my days alone. I have two kids now.
reply