"They ditched a perfectly working MySQL application and migrated to Oracle which caused them to hire Oracle consultants $2000 per day and spend millions on Oracle big-iron."
Reminds me of when the ads manager at Google decided we needed to ditch MySQL and get a "real database" for the ads system. "Real database" in that case meant Sybase, but it was still a barely mitigated disaster. Sometimes it seems as if the "real" databases are just a colossal scam.
Well, it never was listed that Oracle was the reason, just that it was one of, and indicative of why it failed. If the MySQL solution was working, which the answerer on Quora implied, and there was no immediate sign of its failure, I think transitioning to Oracle at the cost of millions of dollars is a pretty big mistake.
Back then it might have been an OK decision, you did get some very nice things with Oracle including scaling and MySQL was a lot less mature, only about half a decade old. It certainly doesn't sound like choosing Oracle per se was instrumental in eToys' failure. On the other hand I've never heard a good thing about Sybase.
Sybase has to be the worst of the "real databases", although SQL Server, DB2 and Oracle are bad too. The first time I tried to use Sybase, I wrote something like "CREATE TABLE foo (id INTEGER PRIMARY KEY, bar INTEGER NOT NULL)". Turns out Sybase doesn't have a type called "INTEGER". What the fuck.
Admittedly, all databases have their quirks. I use SQLite for a multi-process app suite, and get "database is locked" rather frequently. At least my app now knows how to retry transactions. Most of the apps using "real databases" fall over as soon as a transaction fails. "Internal server error, please retry later." Uh, no... transactions are supposed to fail. Even reads.
Oh ye of little history. Do none remember Xanadu still? Inspired by Eric Drexler's call for hypertext as a way of making society more intelligent, they tried to invent the Web, a decade too early and with standards much too high (cached local copies which would still receive micropayments, two-way backlinks), and never solved the incredibly difficult programming problems they were tackling before Mosaic and HTML came along.
Ted Nelson, not Eric Drexler. And not "inspired by" but founded by him. (K. Eric Drexler is nanotechnology, not hypertext.) Not really an Internet startup, certainly one of the longest but not the biggest tech business failures.
Ted Nelson was the founder, yes. He was inspired by Drexler, who like certain later transhumanists did human rationality on the side. Do none still remember "Engines of Creation"? It had a whole chapter on hypertext.
I've worked with Ted on Zigzag and think it's a brilliant piece of software. More people should check out the Java version of Zigzag, Gzz. Basically, where the WWW was Word for the Internet, Zigzag is Excel to the Internet.
It's like Excel in that there are connected 'cells' (called that in Excel and Zigzag), but instead of being in a grid, they can be connected via a link to another cell. A link is either - or +, and the link has a dimension (eg, d.1 or d.name, or d.date). A cell has a string of text as its value.
So say we have 2 cells (which are numbered uniquely), cell 1 and 2. "1+d.1" points to 2, while "2-d.1" points to 1. Cell 1's text is "Here" and cell 2's text is "There".
The final element is that there is a cursor, which sits on a cell, and the program follows all the links from the cursor and draws them onto a canvas, and then shows it.
The user interface lets you make links, break links, edit the text in the cell, delete a cell, and show cells adjacent to the cursor cell. Another aspect is that the screen draws 3 dimensions, and you can toggle the ones shown. This is because showing every dimension is impossible, so we view a subset. It's not a static environment: rather, you make new cells, link them all together, etc.
My work was making a C version. Existing versions were a Perl version (the original prototype) and Gzz (an excellent Java version).
I should have improved Gzz instead of making my own version, but porting it from Perl to C was a good learning experience.
I then made a Lisp version which works great.
Finally, one challenge was getting Ted to accept my innovations and new features, but that isn't easy, so I had to stop working on my version of Zigzag, and maybe even stop using it.
Ted Nelson's Hypercard on steroids? Oh yes. As a bright-eyed technologist of 20 at one of the big 4 accounting firms, I remember being told to 'shut up about hyperspace' In fairness, I had very little idea what I was talking about; Appletalk didn't look like a solid platform and in 1991 few people had a clear concept of the internet to begin with.
I hope I don't find myself looking back in another decade or two and and sighing over the semantic web being too far ahead of its time.
For a time there, I worked for UNext, an online University who decided that it was not enough that their site was powered by Broadvision (using highpriced broadvision consultants), but that their Intranet needed to run on broadvision as well. Oh swell!!
On top of that, they decided to contract those consultants straight out of Broadvision Inc itself, at over $300 an hour.
The CEO drove one of the $100k Mercedes Jeep type SUVs, imported ofcourse.
Not soon after, I left Chicago to go back to the Valley. And they ran out of money soon after.
Great to see Kozmo mentioned a few times, I still to this day have quite a few Kozmo shirts they used to send out for free.
I'd like to add WinStar, they had people going pretty much door to door in Manhattan selling services, I got a pretty decent free lunch off them.
I'd also nominate GovWorks, not huge in terms of money burned or technical blunders, but I credit the movie startup.com for really exposing some sheer ineptitude.
Edit: How could I forget Globix, the colo that had a coffee shop and a gym.
Metricom burned through $600M of Paul Allen's money building out an early wireless network that hardly anyone used, then folded. I don't know if this is quite in the "Internet startup" category since it wasn't a Web site, but it certainly was Internet infrastructure.
I guess technically they don't qualify, but surely Microsoft's web properties have to be right up there. Billions later, they have yet to turn a profit.
They ditched a perfectly working MySQL application and migrated to Oracle which caused them to hire Oracle consultants $2000 per day and spend millions on Oracle big-iron.
They were hiring out of control.
They decided to build a new building to house their new offices because the Ocean Park Blvd offices in Santa Monica weren't pretty enough.
They opened up a London office in Piccadilly Circus, the most expensive rental area in London.
They replicated their entire USA dev team in the UK for no apparent reason.
They started work on a German operation.
They launched a warehouse in Belgium (my baby) to service Europe.
All execs flew first class between Europe and USA.
Engineers were flown around the world as needed.
I was even told to expense my groceries while in Santa Monica for 3 months. Trivial, but the little things add up.
--- none of above would have happened, had they hired professional managers, not a bunch of children willing to spend every single dime on anything other than business itself.
apply that list to any business, including apple, google, microsoft, oracle, groupon, you name it, the result would be the same disaster as with etoys.
In some .com disasters it was actually the "professional managers" who caused a lot of the problems as they were the ones who tried to replicate the infrastructure, pay and perks of the large companies they had worked for before.
how can "professional manager" burn so much money without any profits, knowing that spending this money will not affect the business (positively) -- is really building your own offices because you seen five premises and dont like the view from windows will sell more toys online??
It was a heady time. Success was so inevitable that you could just function on the assumption that you were going to succeed, so you better be ready for it. It was just about to start raining money, the winners would be the ones with the largest buckets, and you better invest in that bucket! Unfortunately what actually rained down turned out to be fire from the heavens, but, well, you know how the story went after that.
"They ditched a perfectly working MySQL application and migrated to Oracle which caused them to hire Oracle consultants $2000 per day and spend millions on Oracle big-iron."
Reminds me of when the ads manager at Google decided we needed to ditch MySQL and get a "real database" for the ads system. "Real database" in that case meant Sybase, but it was still a barely mitigated disaster. Sometimes it seems as if the "real" databases are just a colossal scam.