Never used Supabase before but I'm very much comfortable with their underlying stack. I use a combination of postgres, PostgREST, PLv8 and Auth0 to achieve nearly the same thing.
I feel why the op seems to prefer Miami. It's difficult for those not born and raised in tropical countries to understand the impact that the lack of sunlight has in our lives. We're used to have a good amount of sunlight during the 365 days of the year.
At least the Silicon Valley isn't all bad in that sense...
p.s: I haven't been to Réunion, but I flew over the island when I went to Mauritius. It's indeed tiny but it is extremely beautiful and exotic too \0
Could you expand on that? I'm not familiar with the world of rooting android. Does something exist that would allow you to specify exactly which apps are installed (and remove the web browser and play store)?
Thanks @alexc05!
I deleted the old ones. HN was giving me a damn "submission error" message every time I tried to post, but it turned out that I was actually sending each one successfully.
Pls reach me at felipe.japm@gmail.com or https://www.linkedin.com/in/felipejordaoapm/
teh first question I got from one friend who is planning on stopping by the thread when he gets a chance (but I'll ask on his behalf in case he gets busy) is "why not just use moodle?"
It's the most popular open-source LMS out there and "does all the things" (direct quote)
How would your platform differentiate from that one other than charging for hosting? Why not set up a specialized hosting business called "managed moodle" or something and maybe contribute features back to the mothership?
My main criticism goes to the product itself. The idea of building an LMS came up after talking to several instructors. Although there are many LMSs on the market, many instructors have never considered using Moodle and ended up using paid products. Hotmart and Teachable are examples of paid LMSs that are doing really well.
I think that the added value of those products is that the instructor not only has access to the creation and hosting of his courses, but has also integrated checkout and the ability to create landing pages and promote their courses. My idea is to offer the same product, but for free! The hosted version will only cover our costs. Our business is actually focused on the distribution (https://classpert.com), which we believe is by far the hardest problem to solve.
Now back to Moodle, in fact, I've considered extending it before. They already have a community (including a large directory of plugins) and appears to be used by many instructors and institutions.
However, looking from the outside, I don´t think that extending Moodle will be that easy. I´m not an expert in PHP, but I found the codebase messy. Also, the interface is old, ugly and difficult to use and that is a symptom of bad product to me.
E-learning providers such as Udemy, Skillshare, Pluralsight or Coursera have their own "embedded" learning management system.
But this piece of software can also live alone as some instructors prefer to build their own audience and hence control the distribution (which usually happens through social networks). The advantage over using a marketplace (like Udemy) or selling the content to e-learning providers such as Pluralsight or Linkedin Learning, is that they mainly can sell their courses for much higher prices
Hi guys!
I hope I’m able to contribute by bringing some of my own experience and real data to this discussion. I'm the founder of Classpert (https://classpert.com), a search and comparison site for online courses. In the last 6 months, we’ve managed to sell over 2000 courses, in 8 different languages and across 80 different countries (Udemy alone is selling around 200 courses each month through Classpert). So while it is true that price has an impact on low-income customers (especially from developing countries), even for developed countries (USA, Canada, Germany, Japan) Udemy still is leading the race in number of sales (at least if we use our database as a proxy of the market)
Much of their success stems from the fact that Udemy has by far the largest catalog of online courses on the web (something around 110k courses). And while some people may argue that this comes at a cost of providing low-quality courses it also naturally provides an extremely aggressive long-tail SEO strategy. The majority of potential customers don’t correlate e-learning platforms and quality (most of their customers are not high-profile HN users), so if you are googling for an online course chances are that Udemy will be ranked at the top (and on a global scale). This also explains why they have 10x more traffic than Pluralsight or 3x more than Coursera.
On top of that (an here is much more my personal intuition than data-based analysis), Udemy not only offers cheaper courses but also has not yet adhered to “subscription models”. Subscription models target specific users. Subscription models are awkward and feel totally unnatural to most “normal users”. Why on earth a normal user, seeking for a specific bit of knowledge will lock himself on a subscription? The subscription business model seems to work much better on B2B than B2C.
Good point and I will agree to that; I am pretty much the target audience for Pluralsight, but unless my employer offers it, every time I took a subscription, I felt a strange "pressure" to utilize it (turning learning into chore rather than fun), as well as pressure to unsubscribe unless I can really justify it.
I find it strangely easier to buy a course on either Coursera or Udemy because of seeming lack of pressure :-/
Nice! We're open for feedbacks too!
We received a seed from Quero Education(YC S16). Last year, we’ve made our ways to the finals but eventually got rejected by YC (S19). Not sure if we are trying a second time
NO, because all content is regularly scraped by an in-house crawling engine specially crafted to deal with all kinds of crawling shortcomings! Typically, we crawl hundreds of thousands of courses in 4 days. The offers available through affiliate marketing networks would never work out, they just look like ads
Yes, we currently use affiliate networks to track sales (soon to be changed). This metric ensures that our product works end-to-end.
At this very moment we are growing at 30% MoM and we really don’t care about revenue. We know that the e-learning landscape is completely overloaded and we’re trying to solve this problem on a global scale.
Whoever wants to solve this problem, will have to think big from day one
I couldn't agree more. Over the last 5 years I've changed the way I think about testing. Why would u bother about testing in the beginning when your product is constantly evolving? Also, a senior and organized team helps a lot in this process. I'm currently working with two senior developers and we feel that testing can still be postponed safely until we grow our team or our product becomes more mature and interfaces gets more solid and well defined. Nowadays these are my golden rules for testing a small size project/product:
1 - e2e tests for flows and interfaces that are less likely to change.
2 - write unit tests for core models that are specially difficult to test e2e or core models with complex business rules (if it's a backend code, usually your db table with the highest number of relations)
