3 ms·

Even if it's not the best implementation, it could be considered as scaffolding upon which to improve. While you clearly have enough experience with A* to tell

by AlexC04 15y ago

Even if it's not the best implementation, it could be considered as scaffolding upon which to improve. While you clearly have enough experience with A* to tell the difference between a good and bad implementation, I've not got the same background (I've been reading about it, looking at flash implementations and trying to figure out how the heck I was going to do it in JS+Canvas). Maybe there's a short list of glaring inefficiencies or places where this implementation is behind the curve. What's it missing? I'm sure that for the OP this is a reasonably significant achievement, to me it's voodoo... and to John Carmack it's something else. I doubt Carmack would be so derisive of someone's professional development. Would you tell an 8 year old that their finger paintings were pretty crappy too? Criticism can be valid and encouraging if delivered in the right way.

gradstudent 15y ago
Perhaps we have different expectations of what we think should appear on Hacker News. I come here to find out about cool new stuff that hackers elsewhere in the world are working on. To me, an individual learning a fundamental AI algorithm isn't all that interesting. Particularly as the OP in this case has nothing to add to the discussion; he's just crowing to the world: "Look! I implemented someone else's idea!"
- coderdude 15y ago
  That's retarded. 90% of the articles on the homepage at all times are garbage tech gossip. I don't see you insulting those authors. You're either a troll or just a jerk.
  - gradstudent 15y ago
    Wtf? This is a news site. How is somebody learning something fundamental news? Would you like to see a front-page story every time some kid completes an algorithms assignment? I don't. Not because the learning is unimportant, simply because it's most interesting to the person doing the learning. All I'm saying is this: if you've learned something cool, and you want to tell the world about it, you need to present it in a way such that your learning contributes something to the learning of others. For example: explaining a complicated idea in really simple terms that anyone can understand (such as this cute example of the Halting Problem proof: http://www.lel.ed.ac.uk/~gpullum/loopsnoop.html http://www.lel.ed.ac.uk/~gpullum/loopsnoop.html). Another way to make your learning interesting to other people is to present an implementation that provides insight into the subject. The author tried to do this but I argue his execution is poor. I learned nothing from his app and nothing from his documentation. No implementation insights, no analysis, no beautiful code, nothing. Nada. Zip. Zilch. It's just a web-based A* implementation. Of which there are zillions.

by AlexC04 15y ago

gradstudent 15y ago

Perhaps we have different expectations of what we think should appear on Hacker News. I come here to find out about cool new stuff that hackers elsewhere in the world are working on. To me, an individual learning a fundamental AI algorithm isn't all that interesting. Particularly as the OP in this case has nothing to add to the discussion; he's just crowing to the world: "Look! I implemented someone else's idea!"

coderdude 15y ago
That's retarded. 90% of the articles on the homepage at all times are garbage tech gossip. I don't see you insulting those authors. You're either a troll or just a jerk.
- gradstudent 15y ago
  Wtf? This is a news site. How is somebody learning something fundamental news? Would you like to see a front-page story every time some kid completes an algorithms assignment? I don't. Not because the learning is unimportant, simply because it's most interesting to the person doing the learning. All I'm saying is this: if you've learned something cool, and you want to tell the world about it, you need to present it in a way such that your learning contributes something to the learning of others. For example: explaining a complicated idea in really simple terms that anyone can understand (such as this cute example of the Halting Problem proof: http://www.lel.ed.ac.uk/~gpullum/loopsnoop.html http://www.lel.ed.ac.uk/~gpullum/loopsnoop.html). Another way to make your learning interesting to other people is to present an implementation that provides insight into the subject. The author tried to do this but I argue his execution is poor. I learned nothing from his app and nothing from his documentation. No implementation insights, no analysis, no beautiful code, nothing. Nada. Zip. Zilch. It's just a web-based A* implementation. Of which there are zillions.