nmrm

7 ms·

181.
by nmrm 12y ago
Since others have described the high-level difference between checking and proving, I will point out that you might not want to use a dependently typed language even if you plan on doing some proving. A large cost of formal verification i
182.
by nmrm 12y ago
Here's an interesting blurb about how they were using TLA+. TL;DR is is that, unsurpisingly, model checking trumps proving in roi: We have found formal specification and model-checking to yield high return on investment. In addition t
183.
by nmrm 12y ago
There's a lot to an interview besides just coding questions. Interacting with an actual interviewer lets you get all of the human interactions, talk in-between problems, using problems to showcase talent, etc... all this stuff I'm
184.
by nmrm 12y ago
> Are you literally demanding the candidate spit out the optimal answer immediately? I think the author of the article is indicting situations where this is the case. Perhaps he could have been more explicit up front, but reading betwee
185.
by nmrm 12y ago
> Interesting is fairly subjective Interesting is a social construction. As such, I would define interesting as worthy of publication in a first or second-tier conference at some point in the history of the field. > If you cannot demo
186.
by nmrm 12y ago
> Actually this is a simplification as the chairs slow down and bunch up at the bottom and top to make getting on and of simpler Yeah, this is the problem with the question that I noticed as well. If you don't know what a ski lift i
187.
by nmrm 12y ago
Are you referring to tortoise-hare algorithm? I refuse to believe CS freshmen regularly come up with this on their own . Maybe a few percent, everyone else is googling or relying on huge hints from the lecturer. If you were referring to a
188.
by nmrm 12y ago
If your list is laid-out a contiguously in memory as structs with both next and prev ptrs, you can visit each member using ptr arith and switch the values without actually walking the list using next and prev. Ditto if there's an under
189.
by nmrm 12y ago
I think "coming up with interesting algorithms" was the intended meaning. In which case, no, everyday programming is mostly not about designing algorithms. Otherwise, programmers would spend most of their time with journals and bo
190.
by nmrm 12y ago
> Self-consciously derivative and combinatorial art exacerbates this. You are absolutely entitled to your artistic opinions and I have no interest in challenging those. However, using the hammer of law to enforce your aesthetic perspecti
191.
by nmrm 12y ago
Hm. I understand your individual points, but don't understand the argument you're trying to make. Why does this difference between art and science justify art's comparatively hostile attitude toward "derivative" wor
192.
by nmrm 12y ago
> But a lot of art is original, and speaking of someone's originality is usually one of the highest artistic compliments one can pay. Um, this is exactly true in science. But scientists are generally required to make clear and expli
193.
by nmrm 12y ago
Thanks for the reflection based on your experience. fwiw I spent some time on Oplerno today. Your model (start with credit, then figure out accreditation through existing reputable boards) makes a lot more sense. > ...and as a layperson
194.
by nmrm 12y ago
For me, the social aspect of college was about a lot more than forming a "professional network". I learned a lot and grew as a person interacting with people outside of my area who will probably never (directly) provide me with a
195.
by nmrm 12y ago
You make some good points, but many of your objections are shallow. > It starts with colleges should be more than job training centers, then ends with trying to create an accreditation process to prove value to external entities (of whic
196.
by nmrm 12y ago
Your tone is less than constructive and I agree with the down-votes. But I also have to agree with your assessment given the contents of the article. The bill typifies the sort of "you're so stupid/corrupt/ridiculous we
197.
by nmrm 12y ago
Thanks for replying. these are all interesting points, and I'm intrigued by your prediction. If things work out that way I would be quite happy; pay walls in front of science suck! > - A lot of papers end up in a lot of different of
198.
by nmrm 12y ago
You sound like you're trying to justify a bad decision... Outside of the top-ranked flagship state schools -- which carry as much prestige as an ivy in some fields (e.g. in CS uiuc, u.w., austin, etc.) -- state schools are actually by
199.
by nmrm 12y ago
> I'm not in academia, but when I was in college in the 90's, adjunct sounded like a grad student on a stipend. This attitude was perhaps more appropriate a decade ago, before adjunct faculty were covering a significant portion
200.
by nmrm 12y ago
It's not quite a fair comparison, since a whole bunch of published papers end up on arXiv in some form
201.
by nmrm 12y ago
This sounds like a good description of their approach to the tablet market, actually.
202.
by nmrm 12y ago
Could someone read through the website/bill [1] and verify this story even comes close to accurately portraying the bill? As far as I could find, the text of the bill doesn't even mention the issue of navigation apps. Nor does the
203.
by nmrm 12y ago
> I'm not defending uncompetitive environments and monopolistic exploitation Okay. Well that's what the article is attacking. So I guess I don't understand your point w.r.t. the thesis of the original article. Do you agree
204.
by nmrm 12y ago
Yes, I agree. I was just explaining how it works :)
205.
by nmrm 12y ago
> Value for who? Well they sure as hell aren't passing it on to their employees; meanwhile, their stock is doing quite well... > Wal-Mart delivers value for shareholders because they attract customers because they're cheape
206.
by nmrm 12y ago
> so if I need to check compatibility of types I can just walk the class hierarchy and get an answer. For languages with parametric polymorphism, note that inheritance is kind of like subtyping. Getting subtyping correct in the presence
207.
by nmrm 12y ago
> Consolidation only persists where it is either an effective way to deliver value (Walmart) or regulatory capture has occured (eg taxi licences, car dealerships). Value for whom? The problem is that "effective way to deliver value&
208.
by nmrm 12y ago
angersock, > I imagine that a more inheritance-driven language would be more amenable to strong type checking, for example. AFAIK the conventional wisdom among formal methods people is actually the opposite. If you don't mind my ask
209.
by nmrm 12y ago
Caveat Emptor: All the analysis below pertains mostly to type system features and other "surface observable" aspects of the lanugage. Newly maturing compilation techniques are certainly another reason for the recent explosion (e.g
210.
by nmrm 12y ago
First -- I don't think it's wise or useful to judge Ph.D.'s by the same standards as industrial developers. Or vice versa. Consider the corresponding criticism of industry: "have you meet some of these jokers in industry

More ›