FILL THAT MALNOURISHED MEAT-BUCKET YUO CALL A HEAD, BOB SPAGGOTS

[Elder Iptuous]

Gotcha. i can understand your sentiment.
I would just be remiss to let your statement that software is not engineering go by unchallenged, since i work directly with a lot of coneheads.  particularly since it's in a field where the software and hardware engineering disciplines intermingle significantly.

as far as those acronyms... they just represent a technology that points out the boundary between hardware and software being blurry.

[Nadezhda]

It was more of an encouraging poke saying "Fuck what these morons think! PUT IT IN YOU."

There was an Engineer in my Medieval Archaeology class last month (Summer classes are only 3-5 weeks long!) We had a conversation similar to this about the retardity of some of the humanities students trying to apply their non-logics to our respective group projects, and then he put it in me.  The moral of this story is that we were being snobby Anthropologists and then we had sex.

[Kai]

It was more of an encouraging poke saying "Fuck what these morons think! PUT IT IN YOU."

There was an Engineer in my Medieval Archaeology class last month (Summer classes are only 3-5 weeks long!) We had a conversation similar to this about the retardity of some of the humanities students trying to apply their non-logics to our respective group projects, and then he put it in me.  The moral of this story is that we were being snobby Anthropologists and then we had sex.

[navkat]

I believe that MIT and UC Berkeley both offer more extensive free online courses, so that might be the more effective way to go.

http://ocw.mit.edu/index.htm
http://webcast.berkeley.edu/

WINNAR.

Nigel ALWAYS wins my threads! GOD!

[Mesozoic Mister Nigel]

I believe that MIT and UC Berkeley both offer more extensive free online courses, so that might be the more effective way to go.

http://ocw.mit.edu/index.htm
http://webcast.berkeley.edu/

WINNAR.

Nigel ALWAYS wins my threads! GOD!

[Cain]

If you want to learn how to engineer (as opposed to write code)...well, the Pirate Bay has literally thousands of engineering text books.

Of course, you may need to brush up on your maths a little first....but the Pirate Bay also has thousands of mathematical texts books.

Of course, you may need to  brush up on your reading a little first...but etc etc etc

The only excuse for not having the grand sum of human knowledge inside your skull is a lack of time or interest.

[Elder Iptuous]

i think interactive discussion with a teacher greases the wheels considerably.
or even a group discussion among those learning the topic.
just going it by yourself is significantly more difficult for the majority of people.
i would imagine that since all the knowledge is there, a forum for learning in as efficient way as possible will emerge before too long.  that will be pretty revolutionary..

[Mesozoic Mister Nigel]

i think interactive discussion with a teacher greases the wheels considerably.
or even a group discussion among those learning the topic.
just going it by yourself is significantly more difficult for the majority of people.
i would imagine that since all the knowledge is there, a forum for learning in as efficient way as possible will emerge before too long.  that will be pretty revolutionary..

Speaking as someone who has been almost entirely self-educated since 3rd grade with the help of a bicycle, a bus pass, and a library card, I can say that even the most highly-motivated autodidact may find instructor interaction more conducive to learning in some areas than in others, and peer discussion is really vital in forming a solid grasp on some subjects. I was terrible at math and had no confidence in my abilities until I took a college math class with a real live instructor who explained and demonstrated the principles on a blackboard. My friend Dinah, on the other hand, devours math books alone in her dining room.

I don't need the instructor and the blackboard when I'm reading about biology, and I don't especially feel like I need peer discussion, but in philosophy, literature, and psychology, peer discussion helps develop my understanding from different perspectives.

[navkat]

i think interactive discussion with a teacher greases the wheels considerably.
or even a group discussion among those learning the topic.
just going it by yourself is significantly more difficult for the majority of people.
i would imagine that since all the knowledge is there, a forum for learning in as efficient way as possible will emerge before too long.  that will be pretty revolutionary..

Speaking as someone who has been almost entirely self-educated since 3rd grade with the help of a bicycle, a bus pass, and a library card, I can say that even the most highly-motivated autodidact may find instructor interaction more conducive to learning in some areas than in others, and peer discussion is really vital in forming a solid grasp on some subjects. I was terrible at math and had no confidence in my abilities until I took a college math class with a real live instructor who explained and demonstrated the principles on a blackboard. My friend Dinah, on the other hand, devours math books alone in her dining room.

I don't need the instructor and the blackboard when I'm reading about biology, and I don't especially feel like I need peer discussion, but in philosophy, literature, and psychology, peer discussion helps develop my understanding from different perspectives.

Well put.

[Triple Zero]

there's a big difference between "software engineering" and "writing code".

it's probably not "really real engineering for realness" (though nobody defined that yet, ITT), but probably moreso than "social engineering" which is something else entirely and just called "engineering" in a very different, looser, meaning of the word.

"writing code" is what small webdevelopers do, and what computer scientists do (whether they're writing code for biology or physics, doesn't matter), simple one-off stuff, get the task done, crunch the numbers or serve the webpage. computing science code is generally horrible, doesn't matter, it gets the job done (except that it kinda puts a wrench into the whole "repeatable" bit of scientific research, which becomes kinda tough if nobody can or wants to read your code--but that's a whole other discussion).

"software engineering" is a completely different beast. it deals with billing systems for phone companies (you know how complicated all the different subscriptions are? multiply that by millions of clients signing new contracts every month), accounting hours, treatment and medicine for hospitals, things like that. including the parts where the archives of the system worked completely different a year ago, but still keeping the databases not to contradict eachother, and not lose data, and how certain things could be a lot simpler if they didn't have to pass through several layers of management and government regulations. in order to keep all of this even *slightly* in check, there's many protocols for automated testing, manual testing, deployment and development cycles.

it's also (IMVPO) major boring crap that I hope I never have to do (though it pays extremely well), but just "writing code", it is not.

I have no idea what specifically these courses are teaching though, I didn't look. Just saying that software engineering involves a whole lot more than just writing code, it's not even "writing code at a very advanced level and doing it really professionally" because often the code itself is rather simple (there's a lot of it, though), the tough bits is all the stuff I mentioned above.

[Elder Iptuous]

It's three courses: Machine Learning, Intro to AI, and Intro to Databases.

[Triple Zero]

Ok, I've done courses with pretty much the same titles in university.

My very personal opinion :

The first two are "mindcandy", in the sense that they're really cool to learn about, but not particularly useful career-wise, unless you're going into research and/or doing a full Computer Science education next to it.

Intro to AI is probably quite broad and you'll learn about a variety of topics, though probably mostly from the perspective of getting to know what you'd like to delve deeper into later.

Machine Learning was the topic of my Master's research project (that I never finished), and I can talk about it for hours It's a bit more math and statistics oriented, but from a practical point of view. It deals with number crunching and datamining of big datasets in order to make predictions about it. The basic idea is that you feed the computer (Machine) a bucketload of labeled data points, and you want to make the computer learn to label unseen data in a similar way. Spamfilters are based on Machine Learning, you tell it what is spam or not, and after a while it manages to classify it on its own. Some ML algorithms allow you to tweak the false positive and false negative rates, which is probably important for spam.
In quite a few algorithms, Bayesian reasoning is employed, which is an area of interest to some people here on the board, there's bound to be a lecture on this topic in this course, so they might want to check out at least that video or lecture slides once it is put online.

Career-wise, if you were ever thinking about applying at a search engine company, having a completed course on Machine Learning on your CV will definitely work in your favour. Possibly even if the position doesn't immediately involve code. Similar for analytics-type jobs at larger "social" web companies.

Actually, turns out that Intro to AI has two lectures involving Bayesian learning: http://robots.stanford.edu/cs221/schedule.html
Additionally, possibly the lectures on "Planning in Belief Space" and "Games (Adversarial Search)" (that's Game Theory) may be of interest to the LessWrong-fans as well. (from Intro to AI schedule)
Waitwait I got it all wrong, that's the meatspace schedule. The online cyberspace schedule is different and mixes up the lectures a bit: Intro to AI online schedule, seems to cover the same topics, but a bit more condensed .. hm.

Aha, and in the good traditional of (technical?) university course information, the format for the schedule of the ML course is of course presented in a completely different manner: http://www.stanford.edu/class/cs229/ additionally it seems to be the information for the 2010 course, but there is no newer info, typical ... (I'm getting flashbacks here ... ) Nice thing is that the lecture notes and lots of material seems to be already available for download, so you can start right away! A quick glance shows that the ML course gets pretty technical, though, they're not kidding about the requisites of computer programming, linear algebra and probability theory. So that's not for everyone, I'm afraid. (although there's "Section Notes" available for download with review/refreshers about Linear Algebra and Probability. You still need to be able to write and read code--but there's links to tutorials for Octave [opensource version of MATLAB])

These two courses, however, really have nothing to do with Software Engineering. But neither do they claim to be, BTW, that was just navkat mentioning "engineering" in the OP You can just about call it "engineering" in the sense that it's technical topics. Definitely not Software Engineering though.

Intro to databases kind of falls under the category of Software Engineering. In the sense that traditional SE is really more about the design, management and planning of large-scale software projects (seriously more like part Business, part IT). But databases are usually the central or key component in such large projects, and looking at the schedule/lecture titles, it seems that this course deals both with the technical parts of databases as well as the problems you'll encounter in design/planning a large database-based system ("Data management challenges at Facebook & Twitter"), and selecting the right tool (type of DB) for the job.
Even though I understand the absolute necessity of this kind of knowledge for building large IT projects, this topic was for me major boring shit. Though it wasn't taught well at my university, very outdated stuff, this course seems a lot more by the times. Also because of that reason I can't really estimate the level of knowledge required to do this course.

Possibly the important thing to consider about the Intro to Databases course is that, though it may be boring shit, very complex boring shit even, it is also the subject where the really well-paid solid IT jobs are at. It's always in demand, because so much of our society and industry is dependent on IT solutions that make use of gigantic database systems, and they're always going to need dependable, knowledgeable people that can dig into the innards of these systems in order to keep them running. In that sense, don't tell me it's not engineering.

Schedule: http://infolab.stanford.edu/~widom/cs145/index.html


TLDR: if you wanna learn new smart things for fun, cherry-pick some of the cool shit from Intro to AI. The ML course is probably too hard if you can't linear algebra, probability and computer program. The DB course is a subject I'm not very familiar with, the type of thing that can land you a really good job, though I'm not sure if just one course on DB systems is going to cut it.

[Triple Zero]

bump for enrollment has begun:

https://www.ai-class.com/registration/

[navkat]

HAWT.

Can I touch it?