Thread: Why nothing but beginner questions?

Perhaps my perception is skewed, but it seems to me that there are nothing but beginner questions posted recently.

If it was just this one site, I would assume it is because of the large influx of new learners in education, but I don't seem to be able to find any interesting questions anywhere right now. It's all either questions covered in any reasonably well written programming guide, or soliciting opinions.

If I was the sort to rely on foil headwear, I might think there is an extraterrestrial stupid ray hitting the planet right now, tuned mostly to programming and politics, but as funny and as interesting such an explanation would be, I think it is rather extremely unlikely for that to be the reason.

Of course, there is nothing wrong in beginner programmer questions per se. It's just that after answering the same question a few times, it gets uninteresting. I don't know how Laserlight et al. keep their enthusiasm, really.

I'm just wondering, where have all the interesting/hard questions gone?

Stuff like questions about data structure details when implementing a core (as in "central to solving the problem at hand") algorithm; soliciting suggestions as to how to solve a specific type problem; and so on. You know, questions that cannot be answered by a simple quote from your average programming handbook or programming language specification.

I'm sincerely hoping people are still working on such difficult problems, somewhere, and just not discussing them online. I don't have any such myself right now, just a mind-numbing buttload of mundane uninteresting everyday stuff, so finding a bright problem to consider would do good right about now.

Originally Posted by Nominal Animal

Just out of curiosity, have you tried whether wearing a tin-foil beanie changes your opinion on this?

Edited to add:

I'm not trying to be snarky; I'd just like to exclude the possibility of an extraterrestrial stupid ray. I'm semi-serious: looking at the last few weeks political events here in the cold northern europe, such a ray would explain so, so much ...

Sure, occasionally a beginner programmer will blow your mind, by providing a completely new viewpoint to the problem you might not have even considered. Unfortunately, I haven't seen those either, in weeks. Months. Even then, the original utterance usually stems from confusing some concepts. (Which does not change the value of the new insight, but is indicative of the rarity.)

The kind of questions prevalent here and elsewhere (StackOverflow sites, for example) are those that are best answered by quoting either the programming language specification or standard, or a well-written programming manual for that programming language. Nothing new, just people working through the variations of same. Questions involving new solutions seem to have vanished. Where? Why?

Perspective is a profound word, and I appreciate yours (all advanced); now, mine. Without question, ad-nauseum diatribe of basic questions and students seeking code fix with zero personal effort, well, sucks. We (some beginners) get it. Please remember, however, not everyone who studies programming does so in bootcamp, university or geek peer group. Much is learned by studying text, tutorial and problems; but, without some level of feedback from programming community, much is also missed. I use cboard to fill that gap.

All advanced, please remember there is personal and collective value in sharing knowledge, and a beginner question does not always infer incompetence.

TL;DR:

We know.

But, to be worth the effort and interest to "advanced" members too, the site needs "advanced questions" as well. I personally am wondering where the heck such questions are being posed; and slightly afraid that asking such questions has become rarer now because shipping fast and often is more important than technical quality.

---

Originally Posted by Euclid365BCE

Please remember, however, not everyone who studies programming does so in bootcamp, university or geek peer group.

My first programs were copied from magazines, painstakingly, line by line. I learned not to try the program halfway through without saving it the hard way. The only reference I had was the Commodore 64 manual. Later on, after I got my first '286 PC -- a Hyundai clone with a whopping 40 megabyte hard drive, of which 32 was accessible! --, saved up for a modem, to painstakingly download Ralf Brown's Interrupt List.

My first commercial program was a system of generating and embedding a serial number to a program I didn't get the source code to. I had to compare two binaries with different serials to locate the serial number itself.. I replaced the relocation code in the executable files with my own, optimized version -- except that it also regenerated and compared the serial numbers. Not having enough money to buy TASM, I did it using MS-DOS debug, which is a feat I'm still a bit proud. News like this feel more than a bit insulting, because some of us old folks did that over two decades ago, and got paid, too.

All this is just to say, I hear you. (I think I wouldn't be wrong if I wrote, we hear you, but I don't want to put words into the mouths of others oldsters here.)

I personally never recommend any specific programming guides for new learners, because I didn't have any when I started. I did get formal education in programming later on in the University, which is definitely a plus -- I most enjoyed algorithms (although I found Mark Allen Weiss' Algorithm Analysis in C on my own), and although the mathematical analysis involved in algorithm complexity analysis is dull, it has been surprisingly useful on occasion.

Yet, I myself definitely do my best work in a group. A mixed group, including both main genders, with just partially overlapping strengths, seems to be optimal. The way I learned was definitely not "optimal" in any way; it only worked for me, because I was interested and didn't have that much else to do at the time (dark long winters above the Arctic Circle and all).

Having a place to ask, wonder, question freely, is definitely better.

Having such a place, and one where your person -- age, gender, physical detail or looks, skin color, native language, political views -- does not matter, and we can concentrate on the issues at hand, is nothing short of wonderful.

Originally Posted by Euclid365BCE

a beginner question does not always infer incompetence.

Of course not. However, many (not even most, just many) people find it easier to post a question -- often to multiple programming boards -- rather than read through the tutorials to find the answer. This is the reason boards like this require the asker to show their effort first.

Incompetence does not come into the picture at all, really. It is a matter of whether the asker is genuinely trying to learn and find an answer, and spending the requisite effort, or if they're just trying to get their homework/work done by others with the least effort spent by themselves.

(Those who answer, now that's a completely different ball game. Not acknowledging their errors in their statements (or worse yet, silently deleting their post about it), is an especially sore point for me, and gets me furious.)

Originally Posted by Euclid365BCE

All advanced, please remember there is personal and collective value in sharing knowledge

I'm conflicted whether to agree (because the statement is true), disagree (because the personal value in sharing knowledge is a complex question in real life, and would require a bit of game theory to actually hash out), or be offended (because asking someone to remember they stand on the shoulders of giants, insinuates they have forgotten that fact, and are guilty of hubris).

I do love helping others to learn. Obviously, so do others, too; the amount of help that Laserlight and others provide is nothing short of astonishing.

However, for personal reasons, my capability/capacity to help new programmers is very limited. I freely admit, I am a member here mostly because difficult algorithmic/programming questions are like a drug to me; I am addicted. Unlike physical substances, solving such questions bolsters my self-image, and give me energy to do more (especially the day-to-day mundane stuff).

What's more, I get to learn new stuff, too!

I do occasionally help new programmers, but mostly only if I see I might be able to make an impact large enough to make it worthwhile to more people than just the two of us (the asker, and myself). I definitely try to write my answers so that they don't cover just the particular case the asker is asking, but also some of the similar cases, too. And not just show the solution, but also show how, and why, I arrived at the solution.

If there are no intriguing questions, I would be unable to participate here. Not because I don't care, but because it would not be possible for me; instead of being a positive experience, it would be a plain drain on my time and mental resources, with little to no positives.

The moment when a learner truly groks a new concept is glorious, and definitely gives me an energy boost, but it is too rare to occur in the midst of a problem discussion to be a truly balancing factor. Typically, answers and hints here cause the Heureka! moment to occur later on, offline. Not any less important, but it just does not give the feedback I'd need.

Respect or adulation is not a factor for me. My familial and cultural background is such that I perceive humanity "flat"; the conception of one person being "above" or "more important" than another is just ridiculous to me. (Except for children, because they need protection and nurture to grow into fully capable adults; and have unknown potential.)

A simple "thanks" is enough, although a note that describes an Heureka! moment (and shortly outlines it, not just for me but for others, too) is what I really hope for.

(As an aside: A person should not be respected for their knowledge anyway in the typical sense of the word. Knowledge should be tested, questioned, at all times. Logic, and informative, clear, and robust algorithms and logical structures (those that effectively hold against even unexpected questions) are definitely valuable and might warrant respect, though. Respect the work, not the person, is what I'm saying.)

I would bet real money that I am not an exception. I'm weird, sure, but I bet many "advanced" members share some or most of my points above.

Originally Posted by Nominal Animal

A mixed group, including both main genders, with just partially overlapping strengths, seems to be optimal.

o_O
O_o

Originally Posted by phantomotap

ultimate result of years of "gamification" of support forums.

That explains the trend, but not any sudden changes.

Besides, not all of us play the game. Me, I'd happily give away all my "score" at SO to find out the reasons for the six downvotes I've gotten. Yes, I'm pretty sure the answer boils down to "it's part of the game", but there is always the possibility of me having missed something, and I hate to lose an opportunity to learn.

I'm quite sure I'm not special. So, where are the others? Where are the hard/interesting questions?

The goofy reference to the stupid ray is semi-serious, because I can easily see how economic and commercial reasons might drive programmers to actively avoid having to deal with such questions. (A recent thread comes to mind; perhaps there is a planet-wide push for increased "productivity" via "productivity analytics" among programmers, using metrics designed for factories and manual labor, by economists and analysts who lack actual understanding of the subject matter, but can talk dollar signs into eyes in any boardroom. It's not that far from a stupid ray, if you think about it.)

In which case, those questions are not asked at all anymore. I'd hate for that to be true.

Join the CUDA forums. There are all sorts of hard problems there and the research there is usually interesting.

You can also help me write a LAMP to MEAN blog I want to start.

Originally Posted by MutantJohn

Join the CUDA forums.

I am not interested in becoming dependent on a single vendor, thank you.

Originally Posted by MutantJohn

LAMP

Don't get me started, you might regret it.. The entire security scheme with Apache (which is basically the same as every other HTTP server I've seen the sources of) is bass-ackwards. The SuEXEC scheme, in particular, is an insane footgun, security-wise, except you don't know which end the actual bullets will come out of.

I can prove that a simple group-based security scheme on POSIXy systems would immediately eliminate several categories of security risks, without opening any new ones, eliminating all script drop and defacing attacks. (Additional security for database access would require changes to existing gateways, namely to provide the database credentials in a safe manner.)

However, even adding a file ownership test to current scripting languages (I've looked at PHP and Python) seems unsurmountable to the language developers. Pushing a new security scheme to replace SuEXEC (which is, after all, "known to work" and "very sensitive code") in upstream Apache is, well, basically not going to happen, no matter the reasoning.

Security is one of these questions that are thought of as "too hard to solve properly". Especially for software like name servers, web servers, mail servers .. yet we have individuals like DJB who have proven, in practice, that getting them much more secure is possible. See here and here for more about djbdns and qmail and DJB's security guarantee about them.

All this boils down to: If you want to write new content, a blog or instructions or whatever, just buy the service, making sure they have uptime guarantees, and that you can move the site with a moment's notice if you need to. Writing any code yourself will be limited to very weak security, because the entire foundation of the stack is as porous as Swiss cheese; not designed for secure use at all. Without rewriting the stack you're not going to get better security. It's like adding a padlock to a plate-glass door.

Then there is the entire L part of the stack. A lot of the new-ish features provided by the Linux kernel -- like thread-specific identities and capabilities -- that would be very useful to a service daemon, are deliberately not exposed by the GNU C library. To write better service daemons on Linux, we'd need to, well, basically replace the C library with something not quite C-or-POSIX. That's not going to happen without serious effort.

Originally Posted by whiteflags

I'd like a small sample of the glory days of cboard where interesting questions were asked.

Questions that involve an actual problem to be solved, as opposed to completing an exercise. Do you want me to post a list of links to threads?

This summer I was mostly offline in June and July, but in August there were three threads I enjoyed participating in (this, this, and this). In September, I can only think of one interesting thread I participated in here (this one).

Originally Posted by phantomotap

The "SFINAE" question is rich with complexity ... I was asked to leave the forum as a result of arguments from that thread.

I had to look up the acronym at Wikipedia. It's a pretty good summary of the issue, except for the fact that there are compiler (and version) specific variances on the interpretation of the various rules, which makes it quite a complicated issue to get a good grip on.

The result .. it is infuriating to the extreme. Information is precise, but some treat it as a malleable product of a malleable social process. It is wrong on so many levels; it is really not that different to declaring π is exactly 3.14, just because the notion of a transcendental number is offensive to some sensibilities.

Originally Posted by phantomotap

I really haven't felt that things are "the same" here since Adak died.

I've been a member here for less than four years, but I definitely miss Adak. We had quite a few interesting discussions.

I'd like a small sample of the glory days of cboard where interesting questions were asked.

I've been around since '06, and to be honest, the place feels the same. Even after I got here and delved into the forums past... Not only was this site using completely different software, so there are posts that I can never find, but prior to my arrival people seemed less formal than they are now.

I am noticing that no one asks questions related to new standards, or people asking questions will complain about new standards. Whereas way back in dirty C++03 days, people might have asked about something hard to do in C++03. So, it could be a combination of no one learning new things, and some of us being burnt out on questions answered with an older standard.

Or even some of us just not having as much time for the forum as older adults.

I think it's a combo of other sites gaining popularity as well as C and C++ losing popularity that's hurting the site an detracting a lot of the talent and "interesting" questions.