Leaderboard

a little story : This is the famous Myth of Thamus and Theuth, found in the dialogue Phaedrus by the Greek philosopher Plato (written around 370 BCE). In this text, Plato has his teacher Socrates relate an ancient Egyptian anecdote. Here is how the story unfolds: The Invention The Egyptian god Theuth (Thoth), the deity of inventions, visits Thamus, the king (pharaoh) of Egypt. Theuth presents his various creations, including arithmetic, geometry, and astronomy. Finally, he introduces the written word. Theuth is enthusiastic and claims: "This invention will make the Egyptians wiser and improve their memories. It is a potion for both memory and wisdom!" The Pharaoh's Objection King Thamus is not impressed and responds with deep skepticism. He argues that writing will have the exact opposite effect: Memory loss: People will stop training their memories. By relying on external written characters, they will no longer internalize knowledge. The illusion of wisdom: People will read quantities of information without proper instruction. As a result, they will appear knowledgeable but will actually remain ignorant, becoming conceited and difficult to get along with. The Historical Irony Socrates used this story to argue that spoken dialogue is superior to the written word. He believed text is dead; it cannot answer back or clarify itself when misunderstood. The ultimate irony is that Socrates himself never wrote anything down. The only reason we know this story today is because his student, Plato, wrote it down. This ancient debate resurfaces with every technological leap. The same anxieties were voiced about the printing press, calculators, the internet, and now, Artificial Intelligence (AI).

This particular issue has been in play for a long time. Have you ever gone to a food joint, the power is out, and the staff refuse to prepare food for you because they can't ring it up? Like they've never heard of pencil and paper or making change. If you can't do the basic, human version of your job, you'll be left in the dark (sometimes literally) when the AI can't do its job. Worse, you won't recognize when the AI is making a mistake. Without a real-world backup, and the sense to use it, you're at the mercy of the AI and its hallucinations. It's the height of irresponsibility to replace us with something that needs its hand held all the time in case it has a malfunction.

AI , I like to call it Clippy , is here to stay but you are the architect , in the driver seat. Don't blame Clippy if something goes wrong, blame yourself when blindly believing everything it says. It's just a tool but you are responsible at all times. Like guns don't kill people , its people that kill people. Clippy is very good in finding facts and patterns , but it has no real understanding. When coding something big, give total control to Clippy and for sure you end up with lots of code you don't understand nor control anymore. Just do what you always do, create a block or flow diagram and feed it little bits and test everything. Like this weekend , have a portable airco for my sweat-room. Missed (misplaced) a part so I told Clippy and behold, oh you need this and there you find it, So ordered the part (and a day later I found it in a drawer) , but then Clippy said , oh you have simple (dum) airco , would you like to be able to control it with your phone?... sure, what's on your mind?... it gave me 3 options and I went for gold (of course). Paranoid dragon as I am , did some research (after I placed the order) on how IR boosters work and downloaded the manual and found out the IR booster I just ordered only works when airco has a remote with a display and mine only has buttons. So fortunately was able to cancel the order and found another , much cheaper and works with my machine. Now I can only blame myself, not Clippy. It means well , but it can't be trusted blindly. The better the info you feed it , the better result you can get back. At this moment the term AI is found everywhere , many times as a marketing slogan, designed by AI , controlled by AI , AI knows everything... well I can say for certain there is only one person on this whole planet who knows everything and I'm married to her. ok time to for me.

Spoken dialogue passes through the ears to the brain, the written word causes the eyes to glaze over... I used to know phone numbers, worked in a job where I had to phone many people up - and now with mobile phone contact list and calling over the internet using Teams or Skype contact lists... I cannot remember my home phone number! Yes there is some truth in that RLX AI everything... because it is new and trending, not because it is necessary. AI enabled fridges... because they can and not to make the milk in my morning coffee any better.

You’re going to want to use .NET or ObjectARX for something like this. - The routine would need a fast KD-tree so as not to do brute force curve evaluations - AcGeCurveCurveInt3d. This class contains query methods that return intervals of overlap between the two curves. - Other items like TEXT or MText, I would use custom hashing routine to bypass slow O(N^2) comparison loop, in theory speed would be O(1) - you would need a GUI for properties to be excluded. I.e. layer. - Attribute defs are owned by the block table record, while Attribute refs are owned by Block reference. you'll need to define the behavior - You cannot safely run this on a dynamic block, if one of the items that is deleted is part of an action, the block has undefined behavior. I did some work on comparing blocks by hash here [Python] https://www.theswamp.org/index.php?topic=60513 Actually, I would consider prototyping in Python before porting to C++. Python already has a hasher for points. .NET has some hashing stuff too, but I’ve never used it. good luck

AI has been around a long time, LLM since the 1990's, LISP was created for AI programming IIRC. It's a tool just like VLIDE, VS Code, etc. for programming. Problem, just like before, people came on to forums often demanding a code, etc. be written to make their life easier instead of learning something on their own to do it themselves. Most forums used to provide help and occasionally a custom code, etc. just like... That's a long list of requests. So now they demand AI to do their demands, those with some abilities have success, those that have great abilities have great success, etc. I know recently a man in Lee County, FL was arrested for a crime committed in or near Jacksonville, FL due to AI facial recognition, done by a third county's AI program, no second looks, no reasonable investigation, the man had never been to Jacksonville and a valid alibi for the time showing such. That's the bigger problem with AI right now, accepting the results without question. From the comments I see everywhere a great many seem tired of the AI everything.

I’m loving AI, it’s allowed me to do stuff that would take months to write in days. Example, this dark mode project to BricsCAD (https://github.com/CEXT-Dan/BrxDarkMenu) would have taken months to research.I had already done some win32 programming in the past, but very little. In short, you kind of have to know what you’re doing to use AI. I read about the Brown university thing where the professor made the students do their exams in class and most of the class failed. Scary! People need to at least learn foundational level stuff before using AI. What if the cloud is down, or you reach your token max, would you be able to go old school and continue? Or just sit there and twiddle your thumbs. With regards to anti-scraping, I see lots of open-source projects moving off GitHub to other places like Codeberg so their projects aren’t scraped, I guess so big companies that use AI don’t end up getting their code. I’m the opposite, I want AI to train on my code (if its open source)

Due to my workload and a busy weekend, it will be sometime next week before I fix the one I have posted. I have so far a working DXF-DWG, DWG-DXF, and DWG-DWG (change the version) on the Multi-file Batch Convertor using ODBX, I plan to add DGN, SAT, PDF, and maybe more, but those aren't exposed to ODBX AFAIK, but I am not going to use Express Tools. Thanks for the inspiration and help to get back on this. Besides the original DGN batch convertor I found and adapted, I was surprised nobody ever tried to make an improved version before.

I have mixed feelings about Clippy the AI , at least the one from Google , sure its fast and it can give some relevant answers but you have to double check everything. I am trying to create a dcl editor. Wasn't sure where to start so asked Clippy and it gave me some good ideas on where to begin. But as things got complexer it really started to mess things up. And every time you use it you need to explain every thing all over again because at least the free / unregistered version remembers (saves) nothing . But I learned from my mistakes and started to write a very extensive manual so next time I can upload that so we can better pick up from where we started. My biggest concern is that it creates a black box. You ask something , it gives you something back and you paste and test the code , great , moving on to next part. After a while you end up with a bunch of working code until it doesn't any more and that's where the trouble begins and you notice you have lost your grip on the code because Clippy did it all for you. Like your daddy did all your homework for you and you have to take the exam and ...oops , daddy aint around now is he?

There is no doubt that AI is replacing (and will do so even more in the future) the need to interact with something or someone in order to find ideas and the motivation to pursue them. I don’t know which AI tools are the best for programming (I’ve only used Chatgpt so far), but my experience has been positive in terms of how stimulating it is to have someone to discuss ideas with and refine them while solving problems. This was something that, in the past, could only be found in forums like this one. However, the code suggested by Chatgpt is almost always lengthy and often fails. I suppose that will change over time.

I also think that its AI, especially for new users. I mean AI is good at coding or spotting errors and asking an AI often gets you an answer way faster than starting a forum post. For example here is a chart of stack overflows posts