Udacity lays off 125 people in global strategy shift
techcrunch.com
techcrunch.com
Udacity Likes: I loved Udacity's production quality & instructors. I think they are generally awesome in the balance of teaching concept and practical knowledge. They also don't shy away from more complicated topics like autonomous vehicles, InfoSec, AI / ML, etc., which is a unique asset.
Udacity Dislikes: They are placing too large of a bet on the nanodegree concept. There isn't a clear ROI for students - and it's fairly clear that most companies don't really give a hoot about online certifications.
Their classes are too expensive. Both Udacity and Udemy have great Web Dev classes. Udemy's was $10 for 19 hours of content + responses to any questions I had within hours, and gave me everything I wanted + more. Unfortunately, I had to sift through the Bootcamp and Nanodegree madness on Udacity.
They are unaccredited. My company (maybe most companies?) only subsidizes classes that are accredited. I have a $10,000 annual credit and I use it all on university classes. I would definitely consider Udacity nanowhatevers if I could use my credit on them.
Concretely, I found Udacity was simply too expensive so I preferred Coursera as it had great quality at a good price.
Udemy is full of low-quality trash so I don't bother with it.
EdX has some excellent courses that aren't available elsewhere (for example, Plasma Physics) and also has much more reasonable pricing.
I mean, I live in Europe and for the price of an Udacity course I could attend a full brick and mortar course that is much more appreciated by employers.
Udacity's ND used to be amazing, their Self-driving Car ND was the single greatest online course I've ever taken in my life. Sadly, they now dumbed it down, made it 2-term (original was 3-term). There were some issues with other NDs where marketing was not translating to real experience, such as AI ND, where 2nd term was so dumbed down it could be finished in a weekend, or the late changes to Robotics ND 2nd term that removed a real robot from the plan and instead offered academic discount on Jetson TX2. Also, Deep Reinforcement Learning project's scope changed massively for worse (instead of robot walking it turned into moving a robotic arm). Despite these warts, where else could you get intro to bleeding edge frameworks like TensorFlow/Keras/PyTorch 2 years ago on complex real-world problems (self-driving car, 3D object recognition etc.)? They simply had no competition there, but that market might be too small. Coursera/edX is way behind the curve with their academic courses.
Academic courses aren't usually oriented around "bleeding edge" frameworks etc. That's not a judgement, it's just a difference between how university courses are usually constructed and targeted and how a course/book/tutorial/etc. about some current tech is.
Their classes as you say are clear but this doesn't do them justice. They are uniquely clear. There is no other place -- iTunes U, Masters courses, etc -- that reaches their level of clarity. And, as you point out, for the toughest subjects.
If I don't take Udemy, I can generally read the documentation or read through umpteen online tutorials and figure out how to get an app up in Rails/Django/React etc.
If I don't take the Udacity couse on (for instance) AI for trading that information would simply not be available short of a specialized course at one of the top universitys and even then it might not be of the same quality. I studied Machine Learning for my graduate degree and the quality is the same as the best professors there (as it should be! They took the clearest lectures and perfected it before they recorded and edited it).
I don't usually get fired up about online course providers but they have earned it.
So I have only taken Master's courses at two schools, but I have taken many courses at these schools, and a few online.
The professors were from famous institutes and from unfamous ones. It was a top 20 (but not top 10) school. So assuming that it was a wide sample, I think I have some sense of how good the courses are.
You are right in that they might have even clearer teaching at the top 4 (Stanford, MIT, CMU and Berkeley), though some of the teachers I worked with had taught at some of those schools and/or got PhDs there.
My company is really big and really old, and they do pay directly for nanodegrees through Udacity, and nanodegree completion is very much a factor in promotions because management has extremely poor insight into employee performance and this is one of the few verifiable non-trivial boxes you can tick. I don't know how common this is at other companies, but the strategy has worked for Udacity at least once.
For me at least, the Nanodegree wasn't a defining feature for me so much as having very high quality produced videos.
I like Udemy but the content is definitely not quite on the same level. For me it was mostly about getting over the initial learning curve of getting used to something new (interfaces, set-up). As an intro many of the classes are definitely worth the $10-15.
Early on, I asked for help for why my code wasn't working, explaining why I thought it should. My question was handled by the actual face of the course -- the ostensible expert, Richard Kalehoff, and yet he had only a superficial understanding. All his answers were some variant of "I don't know why this isn't working, copy the example more closely." The conversation went roughly:
Me: Why isn't this code working?
Him: Obviously, because you called .setState in multiple places. Didn't you listen to the lectures? It's asynchronous, so you can't guarantee that it happens in the expected order.
Me: Yes, I know, but the docs say that if you chain them as callbacks, you can ensure that they happen in a specific order <link>. Why isn't that working?
Him: I don't know, no one does it that way. I think it's being overwritten.
<several other commenters point out why that can't be the case>
And I paid $800 for that expertise!
If you've signed up for the course, you can see it here:
https://udacity-react.slack.com/messages/C5T90A6TS/convo/C5T...
They can take hours to perfect an explanation and examples and present it to you. That scales.
They cannot give quality feedback to 1000s of students.
I had a similar experience where they switched the sample data in one of the submissions, so the code would not work.
I don't blame the person for not catching that. Had they told me "this should work, but I don't know why it does not", I would have been fine.
Instead they made thing that they thought might work: Use an alternative notation for matrix multiplication which is different in "subtle ways" (it was identical), change the order of the inputs (the functions were commutative, but more importantly they should have known that), and some other nonsensical advice.
I would have respected them simply saying they don't know but actively bad advice is quite damaging.
That includes knowing how to use the single parameter on the most common React function.
But should the person who understands it best be the face? Or should the person who presents it best be the face? The person who understands it best might design the content, or check that it is accurate. The person who is most gifted at intuitive graphics should design the visuals.
I don't think the fact that the face of course is not an expert is a big deal. I do think they should not be answering the advanced questions. However, with 1000s of questions coming in, I don't know how they should solve that issue. They need a quick way of knowing which is the hard question and which is the easy one. (Is that equivalent to the halting problem? I think it might be).
He didn't simply say "I don't know", he confidently faked it until he made it and claimed that he knew the answer, which was false and based on not knowing how a common parameter worked.
Or rather, it seems you agree with me.
Or perhaps we agreed with each other all along but I am not sure why it sounds like you're disagreeing.
To add to your point. $10 comes up to some thing 600 rupees in India. Most students can, and do buy courses when its that cheap.
Accessibility increases reach.
Sure it won’t give you a video guide of X popular framework intro-documentation pages, but it will actually teach you CS for free, and I’d argue, that if you can complete CS50x and CS50x Web, then you’ll never need a site like udacity.
I guess FOMO sells though.
Sure, but that's a tiny minority of people.
If you have the capacity to complete deep CS courses, you can buy a book for $12 and learn React or Angular or any technology you want.
Most people don't have this capacity.
> I truly don’t get how sites like udacity are able to sell thousands of amateurish introductions to various trends
1. People read that some trend gets you jobs.
2. People don't have the discipline or enough foundation to just buy the $12 book and learn trend by themselves.
3. People find a course online that is really basic and thus doesn't tax their mental resources while convincing them they'll be able to get job later in the trend.
4. People get job, or they don't (I bet they don't, since having low learning capacity isn't a good predictor for getting a job in an intellectual field), either way Udacity gets paid.
It's not a bad business model, but eventually it will collapse because customers routinely fail to obtain the key value proposition: tech skills or a job using them.
But this is how you learn new skills, no? People were learning from books for ages.
> Most people don't have this capacity.
If you can't understand written word, how on Earth could I trust the code you write?
Currently buy mine used on AZ most of the time, always pay around $30.
Citation badly needed for the ridiculous assertion that most people don't have the "learning capacity" for "deep CS courses" (whatever that means). The fact that one can reliably teach programming to eight year olds, boot camp graduates can get jobs in the industry with high probability, and just about anyone in any other scientific field can fill a data science position suggests that CS is not such a rarified academic pursuit.
Yes, I have a CS degree.
Problems exist at many levels of complexity and they can be solved in many different ways. Clever architecture often beats clever programming.
More important: Knowing how to work with other people
I don't care how well you can code, if you can't interface with people and know how to switch up your lexicon there's going to be a problem in a lot of teams. I'm not a programmer, but when I'm in interviews for them I look to see how well they can interact & discuss their projects with me without divulging into intense CS discussion.
To me that sounds ridiculous. Especially since you based it their willingness to push themselves further in something like scratch (e.g. something they have no practical nor social reason to push themselves at).
Intrinsic and extrinsic motivational orientations in the classroom: Age differences and academic correlates.
> Consistent with previous research, intrinsic motivation showed a significant linear decrease from 3rd grade through 8th grade and proved positively correlated with children's grades and standardized test scores
Enthusiastic students: A study of motivation in two alternatives to mandatory instruction
> The present study used Self-Determination Theory as a framework for examining age-related changes in motivation for 57 students aged 7-17 years in the context of two alternative educational environments: a home school resource center and a democratically organized school. Students completed the Academic Self-Regulation Questionnaire in order to assess their intrinsic motivation and three types of extrinsic motivation. In stark contrast to the well-replicated negative correlation between age and intrinsic motivation in traditional schools, there was no relationship between age and any of the four motivation subscale scores in the present study. Interpretations and implications of these findings are discussed.
The difference in teaching philosophy of these profs vs the ones who actually teach is the difference between a fashion model school and the army. The latter accepts anyone and trains them to succeed whereas modelling schools only accept good looking people to begin with.
Arrogance crystallized. These kinds of statements give our entire community a bad name.
If a kid wants to learn how react works on a crowded bus and finds a book difficult to follow -- he has a "low learning capacity". What a terrible thing to say. Gross.
Exercises for anyone who thinks this way:
1. Pick an individual sport (running, swimming, biking)
2. Study an elite athlete in that sport
3. Try to get as good as them.
What you'll notice is that, in fact, you can get really really far.
Coding is the same way -- I doubt someone who can't get through a book would ever become Jeff Dean (same way most of us won't be Usain Bolt), but they can become employable and lead a good, meaningful, productive life doing coding!
Brains aren't so different from muscle fibers ... they can grow and learn anything with enough activation.
I said they generally don't because they don't have the stamina, discipline, and curiosity to do so.
Some do. The minority. Most don't.
Respondents are attacking me as if I am saying this is how things should be.
Trust me, I wish it was the opposite. I've been involved in hiring in all my recent roles, and we'd be happy to hire anyone who has enough motivation, discipline, and interest in the field to learn CS and React through books and come work for us.
There's just not many of these.
I didn't make it so. I didn't stand behind people taking Udacity courses and whispered in their ear "give up!" like some sort of comical Satan.
If I could, I'd whisper "Try harder! You can succeed!"
However when I did end up interviewing folks coming out of code bootcamps, I ended up not being able to offer them a job.
Bluntly, most of them seemed to have no real interest in tech. They initially studied something else, were on a different track in life, typically had little interest in tech at best. Then they heard you can make 6 figures in tech no problem, and I met them 4-10 months later, often after they went through some lazy course designed specifically to game my interview process.
Results from recruiting straight out of bootcamp simply do not justify the effort.
In fact they're so bad that despite all our need for talent, we don't hire anyone straight out of bootcamp anymore. We'll gladly be your second job after bootcamp, but never your first.
Because the vast majority who even complete these bootcamps realize the money isn't enough to keep them engaged in a field they don't inherently like - which is why they didn't go to tech in the first place.
My impression is that a large majority of them leave the field after working for a short while on their new job. This is based on reports from friends in the industry + the fact we get a ton of resumes from people who just finished bootcamp, but very few from people who finished bootcamp and then worked in the industry for any amount of time.
All this is for people who signed up and finished a bootcamp, which is already a high bar. The amount of people who bought a "micro-degree" from commercial companies like Udacity is probably far larger, and only a tiny minority of them even completed this relatively easy course (look up MOOC completion ratios), let alone went through a bootcamp which is far more taxing and less accessible.
People will probably attack me like I'm the evil wizard who conjured this unfortunate situation. They're killing the messenger.
With that being said they were all quant majors in undergrad, and actually enjoyed coding.
I've also seen people with non-quant backgrounds land ok jobs doing menial coding (e.g working at an accounting firm on database stuff) ... but they are still making $50k-70k and happy about it... they got there by doing a coding bootcamp.
Those are the two categories I'm talking about. I'm guessing you probably work at a company like Facebook or Google -- and I agree those companies can't find good candidates that meet their bar. I've worked at those places myself.
Right, this is top edge of bootcamp folks, and these guys do get hired and do succeed.
You're talking about folks who completed a tough STEM degree and then switched to tech. These are the folks you know. But these are already pre-selected people. They were already in the top 1% "most likely to succeed" before they even set foot in the bootcamp.
You're missing the 99% who never went to college because they lacked the desire or ability to learn, then drifted into some skill-mill that took their money.
These are the majority of bootcamp graduates.
> I've also seen people with non-quant backgrounds land ok jobs doing menial coding (e.g working at an accounting firm on database stuff)
Ten bucks say these were non-quant college graduates, and probably a good college too. So now we're not talking about the top 1%, but probably top 5%.
What you're considering "menial coding" is just simpler jobs, but not really simple or brain-dead. That stuff gets automated... often by startups staffed by folks who went to Stanford CS.
While this line of thinking has become en vogue, especially to dismiss the notion of IQ entirely, I think doing so causes more harm to individuals. Yes we all have the same malleable brains with the potential of learning, but it just seems like some people come out way ahead, and over the course of a lifetime, that head start becomes insurmountable, in my opinion, and Psychology 100% firmly agrees.
For all the flack the field of psychology gets, one of the most replicated studies in it's history is IQ and it's predictive nature of positive outcomes such as health, school success, job success, mental health, etc
First a quote, and then some sources.
"IQ doesn't necessarily set the limit for what we can do, but it does give us a starting point. And the truth is some people start ahead."
- IQ measured in childhood correlates with important life outcomes, including school success, job success, health, mental health, and longevity. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4170757/
- All psychological traits are partially heritable, including intelligence http://www18.homepage.villanova.edu/diego.fernandezduque/Tea...
- The Heritability of IQ increases with age https://pdfs.semanticscholar.org/7a6b/ad93d88b0158d449881e56...
- One final one, having trouble tracking the source down but particularly relevant here to HN and YComb -- Higher IQ correlates to success of founders
When I say intellectual arrogance ... what I mean is the following:
Many coding jobs are equivalent to being a brick layer (intellectually)
Most CS degrees teach you how to benchpress 500 lbs.
To a large degree you have to be genetically gifted to bench 500 lbs ... but you really don't need that to be a brick layer.
So my point is not that you aren't right... you are. But rather, my point is, there is a lot of menial work to be done that involves code... and not all of it has to be done by high IQ people. As "software eats the world" more and more work will be produced. Coding will become more like addition and subtraction. If you went back to the age of the greeks, they were probably sitting there thinking that no one would ever need to use this stuff except for "High IQ people" like themselves.
Anything remotely as simple and repeatable as "brick laying" gets automated in the industry. Nobody pays the requisite >$100k/yr in the Bay for something that simple.
You complain about "arrogance", but your notion that there's many coding jobs out there that are as mindless as brick laying actually reveals a much sharper bias, and is simply untrue.
> Most CS degrees teach you how to benchpress 500 lbs.
Right... so you and your CS-graduating friends can "benchpress 500 lbs" intellectually, but fortunately most coding jobs are so trivial they're like "brick laying".
Who are you calling arrogant again?
In reality, this dichotomy doesn't exist. There's no upper-class/lower-class chasm in any tech company I know. Everyone is competing for the same jobs.
Maybe now you can understand why someone who never went to college and just completed a 4-10 month course finds it hard to compete with you Stanford kids.
Automation happens, but you still need somone to supervise things.
Brick laying is completely repeatable iteration, requiring no modification or thinking. Anything "directly comparable" to that can be done by a program. That's not a job - that's a short programming project.
More broadly, there's always a set of roles which haven't been automated... just yet.
Being a sysadmin, for example, used to be a job. People used to be able to make pretty good living from managing a small custom set of servers sitting in a room somewhere. It was extremely common, too. Almost every company had some of these people on payroll. Must have seemed like a good, secure career.
It required a ton more learning, skills, and thinking than brick laying. Yet it got completely automated. Now we have devops, which are programmers who write programs to do what sysadmins used to do. Each devop is effectively replacing hundreds or more of those former sysadmins, by leveraging automation.
Also, plenty of system administrators still make a good living out there. Just as their are still companies running 40 year old code. Automation is less about replacing every system administrator as it is allowing companies to have fewer of them. Which eventually means it’s not a full time job and it’s folded into other tasks.
Generally the costs of automation outweigh the benifits at some point and people sit in the middle. Look at a coke bottling plant and their is one manual step where a guy loads the can lids into the machine. It’s just one guy and he gets something like 1,700 cans a minute so it’s just not a big deal.
PS: See 5:00 to 6:00 on this video: https://m.youtube.com/watch?v=nHQXxhd4CHE#fauxfullscreen
Sure, but that's exactly my point.
One devop who is much more capable in programming, leverages his superior skills to automate away 20-50 sysadmins.
That one devop still has a job. In fact, he is probably promoted and makes more money than any of the single sysadmin he replaces, since he his set of skills and abilities is more rare. But that doesn't change the fact that 29-49 lower-skilled tech workers are now out of a job.
All these people went into sysadmin training because someone naively projected very high demand for their relatively limited skillset. Then they got automated away.
The same process is happening relentlessly. Ever more so as the cost of living in places like the Bay skyrockets. Companies employ fewer people, pay each of them more, but expect a very large amount of output from each.
So telling people "hey, you don't have to be smart, or learn much, to have a great career in tech" is simply misleading. Take a guess at the number of sysadmins who followed this advice, and have out of a job since the advent of the cloud.
To inject some reality into these wild numbers being thrown around so casually: 90 IQ is the bottom 25% percentile. Someone with 90 IQ would be noticeably slower than about 75% of the entire population.
Not just "people and you know and worked with". Less intelligent than 75% of everyone in the US.
I really doubt you worked with anyone like that.
When folks talk about "you don't need to go to Stanford to be a programmer", they mean you can be a programmer with maybe 110 IQ (top 25%, which is still not what most of us here would consider "smart") versus 135 IQ (top 1%, which is what top tech is aiming to hire).
Anyway, I think people vastly over estimate IQ’s of people they work with. Look at the number of 17 year olds in the US compare that with the top schools minus smart people not going to top schools. Now consider each school gets a fraction of that and Stanford students really don’t have that high of a median IQ. Extreme outliers pull the average up, but the bottom 50% at standard are just not that impressive.
Very few people had their IQ measured properly.
Either way, I doubt you ever worked with a programmer who really did have 90 IQ. Not to put too fine a point on it: such a person would appear dumb and slow. It will take them a long time to understand anything. They'll be disruptive and a drag on everyone else.
There's a reason top tech spends billions (!) recruiting a highly intelligent workforce.
> Look at the number of 17 year olds in the US compare that with the top schools minus smart people not going to top schools. Now consider each school gets a fraction of that and Stanford students really don’t have that high of a median IQ.
Well, I did the math.
Per the NCES, there were 16.9m undergraduates in 2016. Stanford had about 7,000 undergrads the same year, which includes all departments, including non-STEM like Art and English. Let's say there are 10 schools at Stanford's level (there aren't). That still means schools at that top level will accept only 0.4% of applicants.
That means they could easily pick only candidates with 140 IQ or more.
Of course, they don't, but the IQ at these institutions will still be pretty high.
> I think people vastly over estimate IQ’s of people they work with.
I think you under estimate the IQ of people you work with :)
I'll be surprised if you worked with anyone below 100 IQ (population average).
In fact I'll be pretty surprised if you worked with any programmer below 110 IQ.
That’s way off base. Schools like MIT and Stanfard are comparable for CS students. But High IQ people are not so limited. So, you need to include say Cal Arts which absolutely stomps Standfard for people interested in Animation etc. Standfard might reasonably attract 1% of the very high IQ population if you assume high IQ people go to elite schools. But, as a CS student that took an actual IQ test and was classified as exceptionally gifted aka 160-174 IQ, I can say never had any interest in doing so. IQ tests are relatively uncommon now days, but historically high IQ people where very likely to choose local options and not infrequently avoid collage and join the military etc. There is very little evidence this has changed.
Anyway, 90 IQ is really close enough to average not to appear significantly different in most situations.
Population average? Maybe. Subpopulation of professional programmers average? Less likely.
Granted, it’s a somewhat more abstract skill than most things. But, plenty of young children learn basic programming and brain development means a 90 IQ adult is generally more capable of preforming tasks than a 110 IQ child. Sure, the child learns faster, but that’s of limited long term value.
Honestly, building the same basic CRUD application for some random government agency is just not that difficult. People like that don’t make 200+k, but they make up the bulk of professional programmers out there.
PS: I did say he seemed a little slow for a programmer, but still capable of the job.
There were 162.63 million people employed in the US as of October 2018, according to the BLS: https://data.bls.gov/pdq/SurveyOutputServlet?graph_name=LN_c...
That makes programmers 0.7% of the total working population.
It can easily be seen as an elite field.
Tech companies could theoretically disqualify anyone with an IQ less than 130 and still be left with over x3 times more candidates than working programmer positions in the US.
Any profession that wants to attract 1+ million people simply can’t be that selective.
You keep making claims, then reliable data is presented to refute these claims, which you ignore, and pile up more unsubstantiated claims.
Most recently in this thread, you claimed:
> Their are around 1.17 million programmers in the US, that’s far from an elite field.
I showed by official BLS data that this means only 0.7% of US workers are programmers, and it can thus certainly be considered an "elite field", and employers can certainly limit job offers to high-IQ candidates. You just ignored that and proceeded to make absurd more absurd, patently false claims:
> Any profession that wants to attract 1+ million people simply can’t be that selective.
An absurd claim, demonstrating complete ignorance of statistics.
If I have to choose 1m out of 1.5m, I can't be very selective.
If I have to choose 1m out of a work force of 162m, then I can certainly be very selective.
It no longer makes sense to resume this discussion.
You are confusing theory with practice. In theory some field like investment banking could attract every single one of the smartest people, in practice that does not happen.
Doctors are considered an ‘elite field’ yet their are 950,000 active licensed physicians (ALPs) in the U.S. Lawyers are not nessisarily as presegious and cover 1.3 million jobs.
Sure, a top law firm, research hospital, etc can be selective. But, you don’t see even 1/5 of the smartest people enter any one field. Chess grand masters have higher IQ’s than the average population but that’s in large part because it’s a small group. Medicine, Science, Art, Law, Computers, etc all have very smart people at the top of the field, but the elite don’t represent the minimum. Even the professional world does not have a monopoly as you will find some people with 160+ IQ’s working in fast food or raft guides for various reasons.
skips ad
Until I started looking for courses on uncommon subjects I am interested in. Then I discovered a new world, basically.
I now always keep one eye on sales at Udemy, and will be spending 200€ on a single course in the following day.
Content quality, as always, depends on the course and the author but generally when it comes to IT and related stuff the rating is a good indicator of quality.
Are these not covered by the perma-sales?
Worse for them, I’m a decision-maker on corporate training. 0 chance Udemy gets the nod there with tactics like this.
[0] - I have 42 total courses in my account. I didn’t bother to count how many were paid vs free.
My strategy is to filter for 4.5 or more stars and more than 1000 students.
I‘m not saying all courses below this threshold are bad but for filtering I‘m strict about these numbers. 4.5 is a good threshold because I found that even at 4.4 there are so much more courses that selection is a pain. Udemy conveniently has a filter with exactly this limit.
If a course is over 4.5 I don‘t care about the rating as long as there are more than 1000 students. There is a class of courses which usually have 4.7 or 4.8 stars and few hundered students but are terrible. I guess this where they land with bought students and ratings.
Which Blender course did you take BTW?
If yes, then the price was probably worth it.
I took a Udemy marketing course that regurgitated most of the information I already knew. BUT it included a bit about positioning that I adopted and which landed me more than $10k worth of clients.
I paid $20 for it. Good deal I'd say
https://www.udemy.com/crystal-healing/
EDIT: not sure it's the same course, turns out there's many with similar names on Udemy...
They surely hold the record for most frequent sales events of all time.
I've used both platforms. Udacity is absurdly expensive. I could buy a university course for the price of a Udacity course. The material is pretty decent and having live graders who give you feedback is pretty useful.
In a lot of cases Udemy is a bargain. There is literally ALWAYS a coupon code to get any course for $9.99. While it seems like 95% of the coursework is pretty weak, I have downloaded some truly excellent content. I can't recommend Todd McLeod's Golang courses and Chad Darby's course on Spring highly enough. 10 bucks is an incredible value.
At the end of the day, building portfolio projects with the skills you've learned from Udemy is much more valuable than whatever recruiting help Udacity provides.
In my book Udemy is the superior product.
Am I a computer scientist? No. Would I get a job as a professional programmer? Probably not. But I got way more out of the Udemy course than I paid for, so I'm pretty happy.
I do wish there was less trash on Udemy though, as a novice its hard to sort through the chaff.
Don't sell yourself short.
I agree that some of what they're advertising looks like 'intro to web bodging' and 'learn to mash tables without learning the maths', but there's also Norvig: https://www.udacity.com/course/design-of-computer-programs--...
Granted, I haven't actually paid them any money.
If you don't believe me then just go on youtube and look for some Stanford, MIT etc. course where they uploaded all the lectures. You'll see the first lecture has lots of view but if you scroll through the rest you'll see the retention drops significantly. I've ran some numbers on this and it turns out the biggest drop is actually in the second lecture.
People can learn in many ways, but there’s a sheepskin effect [0] just by paying $49 to prove you end through it and got the button to add to LinkedIn. There are certainly false positives, but these courses are helpful to 1) help users sort out all the EdX vs. wannabes and 2) demonstrate willingness to learn.
Of course I would never use this to try to determine whether someone knows a skill or is competent or is a learning lover, but it’s helpful as one of many filtering criteria.
I also use it as a casual way to recommend training when I care enough about the person that they learn, but not enough that I want to answer every question that comes up with open courseware (eg, what’s a command prompt, why doesn’t this package work with that package, etc).
Also, Udacity is for learning self-driving cars, algorithmic trading, conv nets for object recognition and stuff you wouldn't get elsewhere.
I'm surprised the shift has taken this long. Employers will spend massively more on employee development than employees will.
Employer: $2,000 for a 5 day class? No problem.
Employee out of pocket: $150 for 6 weeks? I wonder if I can just read some tutorials.
It is interesting though, I can imagine it's really hard to hire good talent for many hiring managers because it is difficult to spot who is good or bad. I am sure they wish there was some kind of standard measure like other industries have.
In my experience, having a mostly self paced class with no hard deadlines, then a stack of hard deadlines for projects at the very end results in stressed out students and a ton of poorly done work. Choose a model and stick to it, either its self paced with no defined end date, in which case people will learn at their own pace, or its a classic fixed format course, with a list of due dates published prior to the class even starting.
Edit: Additionally, the refund window seems much shorter than one sees at traditional colleges and there is still no accreditation for the nanodegrees Udacity lists. At accredited institutions, these types of workforce retraining programs have a bit more heft, and set you up to go straight into a decently paid job. It appears Udacity hasn't solved this problem.
When I buy an online course or book, it is often reassuring for me that the course materials are available in an open source repository. This ensures that students can use the course materials in their personal projects, and increases the commonwealth.
[1] https://www.udacity.com/course/front-end-web-developer-nanod...
Until that happens, none of these MOOCs will obtain accreditation from any accreditation organization that is worthwhile. The quality offered by current MOOCs simply do not compare to what you learn from traditional university.
I count eight Master’s degrees. I’m sure you’ll agree Georgia Tech is an excellent university.
https://www.coursera.org/degrees
Ten Master’s degrees including Computer Science from UIUC and a Bachelor’s.
As far as accreditation or doing something better than a traditional university look to lambdaschool.com Zero to employment ready software engineer in eight months, not ready to be an intern, not well, we’re desperate we can train you, but solidly capable of most front end jobs, decent coverage of algorithms, data structures and basic computer science. They’re not MOOC as they’re neither massive nor open but they may be transformative.
The whole point of MOOCs is to delivery high-quality courses for free. If there is some sort of fee type system then it has to be affordable for everyone, even for people in places like rural India.
No, it's not.
It doesn't seem to be for the people that are running MOOC platforms, either.
> MOOCs is about demolishing the ivory tower of education privilege and making high-quality education accessible to all, for free.
It may be defensible to say that MOOCs are about lowering economic barriers to accessing education, but free isn't the goal (most platforms were purely free initially to establish mindshare, but that's a promotional expense not a sustainable business model.) Those that have retained a free option seem to have it largely as a tool for promoting paid certificate/credentialing programs.
On the topic, how does their security nanodegree look? It starts early 2019 and I've been seriously considering it.
Which topics, in particular?
Not because the class material or teaching was bad. On the contrary it was very good.
Rather because they did not let me access the materials after the class was over. They would not let me postpone completion of the class based on my schedule.
It was very much like a real course. I thought it would be more like a course that I could complete on my own schedule if I missed their schedule. I didn't want their certificate. I just wanted to learn the material.
After paying over $600 for the course, it really rubbed me the wrong way that they shut me out of the course lectures and materials. It would not have cost them anything to allow me to learn at my own pace. But they decided to go that way.
One of the points that was really annoying and borderline fraudulent is that their free courses don't operate like that. Those materials are available to you all the time. It's just the courses you pay for where your experience is worse. Go figure.
Anyway, $600 was a cheap way to learn the lesson not to buy a course from them again.
I really think there's a place for a good company to sell courses on emerging technology, and to charge premium prices. They just have to understand that most people need to be able to tackle the material in a self-paced manner. Some of us have jobs and families.
No wonder it now needs "restructuring".
IMHO the big reason for the difference: Udacity original CEO: former professor (Thrun). Coursera original CEO: former president of Yale (Levin). This is not to imply one was smart and one was not; it just shows that the companies' implementations and objectives reflected the backgrounds of the respective CEOs. In the long run I haven't the faintest idea is better.
I do find the Udacity web site hard to follow, but I don't care at all about credentials/certification so might find the "nanodegree" idea more practical for my own purposes (compared to Coursera) if I wanted to take a class.
disclaimer: my gf works for Coursera, but honestly until she started there the only online teaching company I'd heard of was Khan. I learned all the info in the first two paragraphs from her.
Classic faculty.
I should start keeping an eye on Coursera again.
I would suggest they reduce the number of courses they offer and improve quality. I tend to see them as a more "premium" MOOC but I do agree that quality of content is variable depending on course.
Another area they should naturally look at is training for employees in corporations. I really hope Sebastian Thrun can turn things around.
Everyone...let's be honest with ourselves and what we want from these moocs. I don't know about everyone else but what I want is high-quality courses that are online and basically free. Ads are okay and premium features are okay (like live tutoring or a human given lecture). When I say high-quality I want the content to be similar to what you would get from a Harvard, Stanford or other top university in the field in terms of the content. I want to learn the same topics. Don't water it down and make it simple. I want the real deal with the difficulty and all. Once that is accomplished..work on building tools that make these subjects easier to learn.
This is quite a hurdle to build but ultimately this is what we need. We need a free online university that teaches the same material you would learn from a top university. I don't like the watered down MOOCs that udacity, coursera and udemy offer.
What you want then is EdX. Every course is backed by a real-world institution that people have heard of, and they are free to audit, pay only if you want the certificate and/or to support their nonprofit mission
I think this is Khan Academy [0]. It is a great resource that I've been using for quite a while. They even gamify it a little with points and avatars that you can buy. Funny thing is, like a lot of traditional universities, it lags behind on buzz-wordy courses.
Khan Academy is a better educational environment than I've ever experienced.
Unicorns like Udacity have extremely high valuation but are barely making any money.
Most of their spendings goes into Marketing and not actual R&D for education.
I thought that was an hyperbole. Then I checked on Wikipedia: "While not yet profitable as of February 2018, Udacity is valued at over $1B USD having raised $163M USD from noted investors".
It's never about how much money a company can make, it's only about how much someone might pay for that company. At the end of the day, a lot of startup investment is more like trading baseball cards than producing anything of actual value to consumers.
My facebook newsfeed is flooded with their advertisements lol and I don't even use the newsfeed much.
Anyone notice the google trademark has been remove from their android course?
Also, I'd love to hear about your favorite online courses. I took Andrew Ng's Machine Learning and Dan Boneh's Cryptography (both on Coursera) and they were excellent.