roel_v

Lots of lawyers use WordPerfect. Version 5.1. There's plenty of evidence online if you don't believe me, I wouldn't believe me if this was the first time I heard it.

The direction of conversion doesn't matter. If a recipe starts out with 300 gr of flour, and that gets converted into 1 3/4 cups + 2 tbsp, that's useless (or at least silly and cumbersome) for an imperial unit user. The whole recipe needs to be recalibrated to use imperial units, and that conversion needs to take culture (I should probably say 'locale', I'm not sure what the right term is here) into account, so that it would use 'sticks of butter' in the US and 'grams of butter' in Australia. I don't think this conversion can be automated.

My point is - recipes cannot be converted easily between unit systems. There is more to converting a recipe than applying a ratio table.

"No! Handle units! @milk{4%cup} should be trivially convertible to liters"

No! That's silly and it's not how cooking works! This way you get stupid recipes like on American cooking Youtube channels that just run their volumetric measurements through a converter and end up with recipes that call for 234 grams of flour, 13.5 grams of this or that spice and 78 ml of whatever. Recipes are (well, should be) tuned to the measurement system they use, and packaging availability in the area it targets, so that you get recipes that use e.g. 400ml of coconut milk (because that's the most common can size around here), whole numbers in amounts of ingredients etc. Converting between units in cooking and baking is not simply using a unit conversion calculator on your phone!

(and that's not even mentioning that there are UK, Australian and US cups, tablespoons and teaspoons)

Uh no, not in 'Europe'. Where I am in Europe, 'atheism' is being convinced there is no god. Agnosticism is not knowing whether there is a god at all. Not that you're religious but not to a particular god or doctrine like you claim - I've never heard of that concept before, how can you be religious but not knowing what it is you believe in? I'm not much in the know on the exact nomenclature, but your definition of 'agnosticism' is not in any way supported by the way Wikipedia describes it, and the way I've always understood agnosticism is what is called 'apathetic agnosticism' on Wikipedia - and that is, from my perspective, the predominant understanding of it in 'Europe'.

In the court cases, there was communication between Dotcom and van der Kolk that essentially said (paraphrasing here) 'hey, downloads are going down, we should add movie x or album y somehow, people search for those a lot'. There was a lot of evidence (in the form of emails and chat logs) that showed that their 'file sharing' angle was just a front to their actual business which was distributing copyrighted material.

In your example, one of the abstractions is 'what Python does it use'. Because I'm dealing with this very issue right now, distributing a simple Python tool to a Windows end user, and they're on heavily managed government machines and it turns out that their 'python' command invokes a python.exe somewhere on a network drive and for some reason the '-m venv venv' fails, which then causes all the rest to fail as well of course. Now I don't know yet if this network thing is really the root cause, I've just received some error logs that showed this this morning, so I'm on another round of trying to remote debug this, via email, with someone who is (while patient and willing) not a Python programmer.

Python packaging is such a ridiculous PITA, it's laughable if it wasn't so sad. I tried nuitka as suggested upthread, and send my customer a single binary to try, let's see if it works.

I'd argue that a 'conspiracy factor' of 1 is not a conspiracy, just like a road with an occupancy factor of 0 doesn't have any cars on it, and I presume this was exactly the point of the GP.

Slightly related: I love how in the restaurant of the Lego museum in Billund you order by putting some bricks together in a special format (each dish had it own brick configuration IIRC - maybe it also had something to do with the colors) and then you ordered by having your mini-build scanned by a machine.

Plus, restaurants can practice the same recipe again and again for months or years on end. It would be pretty damning if they wouldn't be better than home cooks after that much repetition.

Without subsidies none of those would survive, especially the French ones, who are notorious for their 'traditional' (i.e., inefficient) farming practices.

You mean 'substance' as in 'objective fact'? Well there isn't, that was the whole point of the comment. Everything as far as I can find is factually correct, and yet how it's all woven together makes the narrative feel robotic and off tune. Like, another commenter brought up that it may have been ML generated, which I don't think it is; but it does give me the same feeling as, say, 95 to 99th percentile ML images - it makes you do a second take for no (prima facie) apparent reason.

Then again, maybe I'm just a nitpicking weirdo. Although I've gotten enough upvotes by now that it seems that others have the same feeling. Maybe someone else can explain the reason for their subjective emotions upon reading this better.

I guess my point on that was 'Landsheer van de Nederlanden' should today not be summarily translated as 'Lord of the Netherlands' since 'the Netherlands' has a current meaning, as opposed to in Dutch, where 'de Nederlanden' and 'Nederland' are immediately and obviously referring to political entities at different points in time, and with somewhat overlapping but still significantly different areas.

Then again, I do know that Emperor Charles is referred to as 'Lord of the Netherlands' in English language history books, and I wouldn't easily know how to do it differently, but somehow there it's always more... padded in context?

Likewise (although it may be the same issue underneath...), equating the area of the Dutch Republic with modern-day Netherlands (which I assume he's referring to but didn't want to use that because of confusion with the previous point?) is... rather simplistic. I mean, I guess it's something someone who's never left the Randstand wouldn't think twice about, but that doesn't make it true.

But as I said, it's not like there are any factual errors I could find, and in any other writing I would have easily understood how difficult it is to convey all that nuance in a piece like this where, frankly, that nuance doesn't even matter. It's just that the combination of subtexts made the whole thing feel weird. To me. Maybe it didn't to you.

Haha I didn't want to go into accusations like that, since it does list an author at the bottom and this is apparently an excerpt from a book, but still.

Meta: it's kind of eerie to read an article like this that is obviously written by an 'outsider'. Although non of what's written is factually wrong (as far as I can tell, I'm no historian), the specific way certain nuances are skipped over gives it an almost 'uncanny valley' feeling. Part of it is that it's written in English and certain nuances just don't exist (like his mentioning 'Lord of the Netherlands' vs 'de Nederlanden/Nederland', his strange 'Aux quatre vents' translation - yes I know it's correct literally, but it just feels so off compared to the idiomatic meaning), others skip over so much historical and current geo-political context that it seems tone-deaf to sensitivities ('the Dutch Republic (loosely modern Holland)'). I understand that for an article on an historical painter, you cannot go into details like this, what I'm saying is the choice of what the writer considered important and the way he skips over nuance makes it feel very alien.

I guess this phenomenon is well-known by people who are from cultures that are usually written about by outsiders :) but I don't encounter it often and this specific article just evoked this really weird feeling in me. So many instances where I internally reacted with 'WFT is this guy saying' yet no actual factual errors that I can find.

I was cleaning out my old books just a few weeks ago and put 'Alex and Philip's guide to web publishing' in the 'can't justify to keep, feel bad to throw away' pile. I've been trying to give it away on our local ebay but doesn't seem anyone still remembers the cultural significance, or cares about a book on it...

There is no 'meaning' inside these AI's. It's terribly confusing to think about these LLM's as having 'meaning' in the same way we humans do. It's all just statistics. Given a sequence of numbers (each representing some abstract token), what is most likely to come next. That's how 'simple' it is. It's also what makes it so amazing that these things work as well as they do. I giggle like a schoolgirl every time I get it to add some functionality to a function, or write an entire new function, and that's several times a day for what is now months on end. But the key to using them is seeing that there is no 'meaning' in them. It's all just streams of (to the machine) meaningless tokens.

"Also gpt4 >> gpt3.5 for coding. "

This is not my subjective experience. It's better, sometimes somewhat, but also much slower. I'm not sure which one is the better tradeoff for real work. I have it on 4 by default now, don't want to think about this for every question I ask, but for my work, 3.5 was fine.

I don't think this is the take away. I'm in a similar situation as the GP, but the crux is that we still need people to make the decisions on what needs to happen - the computer 'just' helps with the 'how'. You need to be a domain expert to be able to ask the right questions and pick out the useful parts from the results.

I'm also not so worried about 'oh but the machines will keep getting better'. I mean, they will, but the above will still remain true, at least until we get to the point where the machines start to make and drive decisions on their own. Which we'll get to, but by that point, I/we will have bigger problems than computers being better at programming than I am.

No, that's how it is. My MO is now I type in the question to the chat; go to another tab to google a bit on the background of the problem; by the time ChatGPT is done, I already know enough to have a reasonable idea about whether what it came up with is in the general direction of where I want to go. And if it is, I'm usually confident enough that it gets the details then mostly right.

One meta comment is that I would love for something like a Greasemonkey script that can automatically add some checkboxes or something that will then automatically add stock phrases to ChatGPT questions, like 'show me only the code', as I find that a lot of the intro fluff and the explanation of things are useless to me and just take up time. In general, the UI is pretty bare bones, there's a world to be gained there.

Make it write a test suite that you can manually verify/expand as necessary, just like code written by humans. I don't understand this 'but how do you know it works' argument. How do you know the code you wrote yourself works? I hope you don't rely on your superior programming skills to just get it right the first time?

"Good indoor lightning"

Man I'd love me some indoor lightning, no matter the cost.

"any old x" is an idiom that references "any random" or "any", indicating x being interchangeable or not special in any way. It has nothing to do with age.

Yah 'supposed', but that's (to me) in the same bucket as 'different features behind flags aren't supposed to interact' and 'we're supposed to have detailed documentation of who is/was using what set of features at time x'. That's also what I meant with 'it's all in the process' - if there is good management of all of this, it may work. And I'm sure that for some people, it does actually work. Meanwhile for me, I'm staring at build scripts with 15+ year old feature flags that at this point are realistically never going to be removed, and people still sometimes suggest 'yeah but this one little thing we can put in a custom build until it works for everyone else too...' So now not having feature flags is a hill I'll die on.

Maybe this works in the type of work this guy does, but I've been horribly burned by feature flags. How do you test for the combinatorial explosion of feature interactions? Even if you separate into 'silos' that pinky promise do not affect anything beyond some 'defined' boundary, interactions are still too numerous to test as different products very quickly. And now you have person A using 'feature flag combination' A' and person B using feature flag combination B', and they ask you a question about how this or that is supposed to work, and you won't even know until you are able to exhaustively reason through what the effect of every feature flag this person has en/disabled is - if you even know that at all.

Yes in the end it's all in the process, but feature flags are not the end-all be-all these few paragraphs posit them to be.

I tried that last time, and got upsold a bunch of 'variations', and looking back on everything I bought the next day, none of them looked all that great. The style was too much... well let's say 'country' (which is an accurate reflection of where I am, but I think a professional photographer should be able to adjust to a client's wishes, and I told her specifically what look I did not want - my market is not local), and the studio didn't even have a few colors of shirts I could try on, nor could she advise on the most basic of styling. I mean, they don't even need whole shirts, just get a few collar/shoulder panels for people to throw on ffs.

I'm soured on the whole 'support local' thing (not only because of this episode). Too many chumps.

Not sure why I'm downvoted - I literally used it last week. I got an API key from my account and made several calls to it, using the OpenAI Python library.

Yes: https://platform.openai.com/docs/guides/chat . I wrote a small proof of concept plugin for vim last week, was quite easy to do with their Python library to access this API. I have no experience accessing the raw endpoints.

Oh I don't know about that, I can only comment on my own experiences. I don't have sufficient breadth of experience to be able to say how much of this would be applicable to other developers in other subfields. I do however, like you, look forward to multi modal capabilities. I fully expect it to be able to take a Balsamiq mockup and generate the code to set up the UI for that mockup, with handlers for widget events and all. I've already had it generate UI's from text descriptions, which was cumbersome (relatively speaking - hedonistic adaptation strikes hard here).

Wrt to productivity - I do think it's misleading to look at it like 'it used to take me 5 hours, now only 1, so that's 5x' or some other (imo naive) quantification. It's much more about removing drudgery and thus freeing up mental resources for interesting things. Say I do a task with a certain tool in the same time it took me without it, but in a much less mentally taxing way (because that's what drudgery does to me - activate bore-out mode). From the naive metric pov there is no productivity improvement. But if that leaves me capable of doing more in the rest of the day/week because I don't have to fight myself to move on with something, I'd argue it's still a productivity gain.

Does typing with a nice keyboard and in a good chair make me more 'productive'? Well, I don't think anyone will argue that I can't type on a e5 Chinese crap one while sitting on a bucket before a door on two sawhorses. I can even type and think equally fast in those circumstances. But better tools still make one more 'productive' in the larger picture, I don't think anyone with experience will deny this either. This 'productivity gain' is in that sense somewhat ethereal, in a poetic sense of that word. To me that's where these tools are at now.

Well... yes? I don't know about this 'flow based programming' per se, but what you describe is indeed how one uses ChatGPT for programming. I don't understand the pushback on every article like this. How aren't there obvious major productivity gains when a computer can generate 80% or more of your solution, and you just have to fine tune it?

Let me give you an example of what I did just this morning. In fact, ChatGPT is generating code in another tab as I'm typing this - GPT4 is quite slow (well, at least it takes longer than my attention span which has been reduced to that of a sea urchin by years of social media and usage of internet in general), especially when you let it create 50 to 100 lines at a time.

I'm working on a specialized ETL tool for geospatial data. In fact, I've had several iterations of this tool over 5 or more years, but for the last few months I've been working on a new version that incorporates everything I've learned about the domain in that time. This conceptualization, the learning of what the tool should do and how all components fit together, is not something a computer can do. I would have to be able to describe it to get a computer to generate it, of course, but I couldn't a few years ago because I was still finding out what it should do to provide the most value. The reason I had several versions is because I kept learning more about what I need it to do.

But many of the boring machinations I absolutely can have a computer do. For example, a part of this tool is a tiny DSL which is basically a way of specifying combinations of parameters with some dress-up. I'd been dreading making a proper parser for weeks, since a) the 'design' of this DSL has fluctuated quite a bit, and b) I'd have to dreg up my knowledge of parsers from the last time I wrote one 10 years ago. So I've been getting by using a combination of regex, split() and some hand massaging of strings.

But this morning, I had ChatGPT write me the biggest part of a parser using three questions:

- "Write a parser in python for the following format: [optional identifier] [open square braces] [either a list of values, separated by commas; or a list of key-value pairs, separated by commas, the keys and values are separAted by equals signs] [close square braces]. The identifier is optionally followed by an 'at' sign, and a qualifier after that. Show me the EBNF and a parser in python that implements this grammar."

- "The qualifier should come before the square brackets, it's part of the identifier."

- "Write some unit tests for this grammar. Include corner cases for all varieties of the allowed input, so with or without identifier, with or without qualifier, with plain lists and with lists of key/value pairs. Include cases for 0, 1 and 3 values in all places where more than one element can appear."

(try it, I used GPT4 but you'll probably get similar results if you use the free version)

The result I got from this has 2 mistakes: in one case I had to swap two tokens in the EBNF. This one I found by reading the docs while waiting for the output of whatever ChatGPT suggested as a fix after me copy and pasting the error message I got. So this was fixed, by me, in a few minutes (ChatGPT's suggestion for a fix made no sense). The second one, ChatGPT suggested a correct fix after I copy and pasted the error message also.

So, ChatGPT gave me a skeleton grammar, complete with the syntax of the library which I had never heard about 2 hours ago. The second of my first three questions was because I gave it ambiguous input, I only thought of that part while I was writing the question and didn't bother to go back and edit it in in the right place where it would have been more obvious. Then I copy/pasted the raw output from ChatGPT in a new file, ran it, I got the issues fixed (in collab with ChatGPT) in 20 minutes or so. This would have taken me all morning and probably all day if I include all unit tests and all other details just a few months ago. How is this not a massive game changer?

Now, I would not have been able to do any of this hadn't I already known what an 'EBNF grammar' was, and hadn't known how to read and implement one. This tool will not replace programmers. It will however eliminate the need for the very low skill ones, and make the rest (much) more productive.

No offense mate but if your only experience so far is having an 8 month old, you don't get to have an opinion on this topic yet.