4 ms·

Most of the time. Take a look at developer tools for instance: jenkins,nagios,graphite,jmeter,mysql,postgresql,puppet,vagrant,virtualbox,mongodb,npm,docker,ope

Most of the time. Take a look at developer tools for instance: jenkins,nagios,graphite,jmeter,mysql,postgresql,puppet,vagrant,virtualbox,mongodb,npm,docker,openvpn. What all these software have in common. They are the most popular despite being poor products and often having notably superior competitors.

sho 9y ago
Your list makes no sense to me. PostgreSQL is a poor product with noticeably superior competitors? This is - how should I put it? - a niche opinion. And besides, popularity is its own quality sometimes. NPM might not be the "best" but everything's on it so it's the most useful. A lot of devops people already know docker. If you need to hire, there's a deep pool to draw from. That counts for a lot.
- user5994461 9y ago
  Obviously, the list doesn't make sense unless you know the products mentioned and their competitors, preferably with a bit the history over the last decade(s). The only thing all the products mentioned have in common is that they are free to use, as-in no upfront money. There are pools to hire for any software. First, the hype is cyclical, people who worked long enough will know multiple products. Second, new comers won't know anything so they will need to be trained anyway.
- jandrese 9y ago
  I'm guessing he's referring to Oracle as the competitor.
  - sho 9y ago
    Well that's unexpected! I'm aware that Oracle has a few features pg doesn't, and replication is supposed to be easier, but at extraordinary cost and with the added bonus of supporting one of the worst companies in tech. I don't really think you can even compare them as they serve totally different markets. If I was the CTO at a Fortune 500 and needed a DB for my new global logistics system, fine, Oracle would be in the running. Since this is HN though, I was thinking in terms of startups - and a startup choosing Oracle would be pure insanity unless they had an amazingly good reason.
jandrese 9y ago
Your list is heavily loaded with free products that are good enough. The lesson is that if a commercial product wants to be widely adopted it needs to be significantly better than the similar free products to compensate for the hassles and headaches of buying software (and the god damn support contracts). Simply being free is a major advantage right off the bat.
- user5994461 9y ago
  Developers would rather die than spend a dollar on any product, no matter how better it is or how bad they will suffer. That's only half the lesson though, because some of the alternatives are free or free up to N users. The other half of the lesson is branding and hype cycle.
  - sho 9y ago
    > Developers would rather die than spend a dollar on any product That's just silly. I'm a developer and I have spent plenty of money on good products that help my work. Hell, literally today I dropped $30 on a word processor. And there is a huge market for developer-focussed SAAS and PAAS products. > no matter how better it is or how bad they will suffer Example? Citation needed. > The other half of the lesson is branding and hype cycle These cryptic pronouncements are kind of useless. Again, give an example. Out of those things on your list I'd give you mongodb as a hype-driven mistake but that's pretty much it.

sho 9y ago

Your list makes no sense to me. PostgreSQL is a poor product with noticeably superior competitors? This is - how should I put it? - a niche opinion. And besides, popularity is its own quality sometimes. NPM might not be the "best" but everything's on it so it's the most useful. A lot of devops people already know docker. If you need to hire, there's a deep pool to draw from. That counts for a lot.

user5994461 9y ago
Obviously, the list doesn't make sense unless you know the products mentioned and their competitors, preferably with a bit the history over the last decade(s). The only thing all the products mentioned have in common is that they are free to use, as-in no upfront money. There are pools to hire for any software. First, the hype is cyclical, people who worked long enough will know multiple products. Second, new comers won't know anything so they will need to be trained anyway.
jandrese 9y ago
I'm guessing he's referring to Oracle as the competitor.
- sho 9y ago
  Well that's unexpected! I'm aware that Oracle has a few features pg doesn't, and replication is supposed to be easier, but at extraordinary cost and with the added bonus of supporting one of the worst companies in tech. I don't really think you can even compare them as they serve totally different markets. If I was the CTO at a Fortune 500 and needed a DB for my new global logistics system, fine, Oracle would be in the running. Since this is HN though, I was thinking in terms of startups - and a startup choosing Oracle would be pure insanity unless they had an amazingly good reason.

jandrese 9y ago

Your list is heavily loaded with free products that are good enough. The lesson is that if a commercial product wants to be widely adopted it needs to be significantly better than the similar free products to compensate for the hassles and headaches of buying software (and the god damn support contracts). Simply being free is a major advantage right off the bat.

user5994461 9y ago
Developers would rather die than spend a dollar on any product, no matter how better it is or how bad they will suffer. That's only half the lesson though, because some of the alternatives are free or free up to N users. The other half of the lesson is branding and hype cycle.
- sho 9y ago
  > Developers would rather die than spend a dollar on any product That's just silly. I'm a developer and I have spent plenty of money on good products that help my work. Hell, literally today I dropped $30 on a word processor. And there is a huge market for developer-focussed SAAS and PAAS products. > no matter how better it is or how bad they will suffer Example? Citation needed. > The other half of the lesson is branding and hype cycle These cryptic pronouncements are kind of useless. Again, give an example. Out of those things on your list I'd give you mongodb as a hype-driven mistake but that's pretty much it.