[ex_ubiquiti]

The early days at Ubiquiti were good. I worked with a lot of good engineers and we shipped good work. The decline is a recent problem.

> How the brand hasn't become toxic already is a mystery to me, yet look at the stock price tracker. It's been trending up for years and it has well over doubled in the past six months alone.

This is your answer. No incentive to change. All of the bad engineering decisions have been rewarded by increasing stock price and continued sales.

Most of the original engineers have quit by now. I lost track of how many UniFi engineering leads joined and then quit after it started falling apart. Before I quit, I heard rumors that the CEO was making two separate teams work on the Dream Machine project separately, competing against each other. That made more people quit. I think they were trying to reboot engineering in foreign countries when I left because it felt like we were forgotten in the US offices.

[gotathingtosay]

>This is your answer. No incentive to change. All of the bad engineering decisions have been rewarded by increasing stock price and continued sales.

It'll come around, it just takes waaaaaaaay longer than you'd think for a slump in engineering quality to be reflected in the market. Especially with hardware.

We have a few publicly traded clients that we've worked with for decades (and by "decades" I mean longer than I've been alive). It's cyclical that they want our engineering to build new products when they're doing bad in the market, and once our work is released and gets them some success they'll design transfer back inhouse as aggressively as possible (their engineers aren't all bad, it's just not an engineering culture there). By the time we're out, they're still riding the upswing. Their management's institutional memory either doesn't see the cycle and/or they don't care beyond the next few quarterly reports.

What I'm trying to say is I know hurts to see your baby languish but it catches up to them, eventually.

[Loughla]

>I heard rumors that the CEO was making two separate teams work [. . .] separately, competing against each other.

I don't work in tech, so maybe I'm dumb to this, but why would you ever do this?

[_pplp]

This is not surprising to me at all.

IMO, the CEO had a bit of a Steve Jobs hero-worship complex, but only all the bad parts. I can absolutely see him putting two teams on the same project, and "may the best product win".

The team that "lost" would get canned, obviously (I saw it happen to two separate offices while I was there).

[tablespoon]

> IMO, the CEO had a bit of a Steve Jobs hero-worship complex, but only all the bad parts.

Part of me wishes Steve Jobs had never been brought back to Apple and died in obscurity. He's such a bad example. People idolize him, but his good parts can't be imitated, his bad parts can, and a lot of people can't seem to tell the difference.

[letitbeirie]

> his good parts can't be imitated

Without dropping acid at work at least, but that seems to be frowned upon these days.

[gralx]

Intel tried this too, according to an ex-Intel employee here. It's a management strategy intended to get the best result by inspiring competition. The problems it invites are the obvious, but the tradeoff may be justified in some scenarios.

It's also the premise of David Mamet's famous play Glengarry Glen Ross.

[g_p]

Google certainly seems to do this when it comes to chat applications. Ironically though, they've actually (arguably) lost marketshare - they went from gtalk being pretty widely used (in the late 2000s, early 2010s, as Android took off), to having a confused and fragmented ecosystem (Allo, Duo, Hangouts, Chat, Messaging), and it seems none of those have the same market penetration as the original did.

Perhaps internal competition to that extent simply confuses customers?

[takeda]

They essentially destroyed all competition (AIM, YIM, ICQ, MSN etc), the open source solution that would standardize chat (XMPP) and themselves. Making people just go and use proprietary solution like WhatsUp.

XMPP was so promising.

[Zash]

Psst, hey, XMPP isn't quite dead yet! Some of us never stopped working on it. Come help bring it back into the hype!

[aspaceman]

There’s an infamous anecdote with Jobs doing this. Tharanos had the same “two teams” story.

A lot of CEOs who think they’re the next Steve Jobs, don’t understand their own tech, and presume the solution to their technical problems is a lack of “motivation”.

Creating a skilled skunk works team to handle a critical problem is a great idea. Making two? And putting them in conflict? It’s like throwing your a steak to your dogs to have them fight over dinner. Idiocy.

[roland35]

I can see why the idea is tempting, ie testing multiple strategies and survival of the fittest. But in reality there are extreme downsides. Teams will lie and fudge data to get ahead. People dont trust their coworkers.

I think this is where strong technical leadership is needed. At some point someone needs to make a decision on the technical direction and have the conviction to stick with it.

[jakeva]

I imagine it comes from some flawed business belief in the survival of the fittest. I've never heard a tech person advocate for it, I only ever hear it from business types.

[Silhouette]

Of the things I've seen reportedly happening at Ubiquiti, that one makes more sense than some.

Businesses put projects out to tender all the time, and other businesses that can provide what is wanted invest sometimes very considerable resources into putting in a bid, knowing that if they don't make the winning bid then those resources will mostly likely be completely wasted. Evidently it is still worth operating a business on that basis because the benefits when you do win outweigh the costs of the failed bids, and those costs might include reducing morale in a team who worked on a failed bid.

If that is the case across industries as a whole then economically it might make sense for a business to operate on the same basis internally for their Next Big Thing. Run multiple independent teams at the start, give them all the same brief, then see which team comes up with the most promising starting point. I don't see much of an argument for continuing the internal competition beyond the concept to prototype stage, though, unless perhaps it turned out that more than one team could produce a product that was viable in its own right without competing for the same market.

[fletchowns]

Isn't Oracle notorious for doing this?

[ihsw]

What do you suggest for someone leaning on an EdgeRouter Lite (with EdgeOS v1.10.11, staying far away from v2.x) and a Unifi UAP-AC-PRO access point?

The router will probably reliably carry me until saturating 1Gbps becomes a daily occurrence and the access point will be retired when WiFi 6E comes around (assuming Ubiquiti's WiFi 6E access points aren't required to connect to the cloud.)

[mafro]

Also in answer to sibling comments - you don't need to connect the UI software to the cloud. I have an Edgerouter SFP-X and a few AP lites. I recently added an 8 port Unifi switch for more PoE ports.

Following is to the best of my knowledge! Any ex-Unifi folks or other pros are welcome to correct me:

- The Edgerouter absolutely does not talk to ui.com (except check-for-updates). There's no remote control ability etc etc.

- The Unifi range can be controlled from the cloud, but via your Unifi Cloud Key. You can run this software yourself, without buying extra hardware. When it is not running there is no comms to the cloud. Run the software, configure things, stop the software - I run it in docker on an rpi4.

[severino]

Can you please elaborate more about the thing with EdgeOS 2.x? I haven't been following the news about Ubiquiti until now.

[gbrown]

What’s the issue with newer EdgeOS? I thought the cloud crap could be disabled.

[owly]

I have the same setup and question. Anyone?