On Arista

Investment Analysis Clubs Saul’s Investing Discussions
1

As you know, a very preliminary decision in the Cisco suit went against Arista about a week ago.

At the end of last week Arista implied to Needham that if they lost some of the patent suits it would be no big deal and they could work around it. People on the MF Arista discussion board were rather excited and encouraged by that. I asked, realistically, what else can management say? Can they admit that losing the lawsuit would be catastrophic, and thus:

Scare their customers away?
Make it impossible to hire good talent?
Drive their stock price down and make their present employees stock options worthless?
Drive their stock price down and make acquisitions using stock much more dilutive?
Give the general impression of a sinking ship?

They HAVE TO say that it would be no big deal, even if they didn’t believe it a bit!

Andy (nevercontent) then implied that all Arista has to do is change some commands: Or MAYBE it just isn’t that big of a deal. So they have to change show running config to show my config now. Or Show Interface to Show Interface Port. Juniper has many commands that they changed, but many that are still the same as Cisco. How can Cisco let one company use the command but not allow another? Now let’s say Arista takes the time to make up new commands that are close to what Cisco has but totally different.

I responded making clear that wasn’t implying that it IS a big deal. I have no way of telling if it’s a big deal or not. I was just saying that they have to SAY that it’s no big deal, whether it is no not! That’s a different issue. It just means you can’t bank on what management is currently saying.

Now Andy is a pretty bright guy, BUT, if you were a purchasing agent, would you buy a system from a company that says they may be almost driven out of business if they lose this big lawsuit. How would you get service or spare parts, or problems solved in the future? It’s just easier and safer for the purchasing agent to buy from someone he or she is sure will be here three years from now, even if the product isn’t quite as good. Or if you were an up-and-coming tech guy, would you take a job at a company that might be wiped out in a year or two? No way!

My contention that they have to SAY that it’s no big deal, whether it is or not a big deal in reality, is a different issue than whether it’s true or not. It just means you can’t bank on what they are currently saying about it being no problem if they lose. THAT seems pretty obvious to me. I don’t know how it will turn out and whether they will win or lose the lawsuits, but I think it would be a big mistake to assume it will be a big nothing IF they lose just because management intimates that.

I strongly doubt that they and Cisco would each be paying millions of dollars a quarter in legal fees if it was just a question of changing some wording of commands, and it was not going to matter much if they won or lost. I figure that this may be closer to an existential threat IF they were to totally lose the suit. That’s a big IF, of course, but I wouldn’t totally discount it considering that they lost this first, very preliminary decision.

Saul

For Knowledgebase for this board
please go to Post #9939.

A link to the Knowledgebase is also at the top of the Announcements column
on the right side of every page on this board

11 Likes
2

The point with the similarity to CISCO CLI commands and management is that the network admin doesn’t have to learn a new set of commands specifically for the Arista switch.
Loosing that is certainly a blow to ARISTA but not a blow that will kill them.
Reading about their products, I saw that they are selling Ethernet switches with the advantage that they have the strongest and fastest switch out there (using Broadcom chip set).
CISCO and Juniper are just behind and will catch up with them.
For me there is not enough moat to protect future earnings from disruptions by competitor. With the new ruling that makes assimilating their switches a bit more complicated I will pass.

5 Likes
3

Andy (nevercontent)

A little typo, Andy is buynholdisdead :wink:

Neil

1 Like
4

Sorry Neil, careless of me.

5

Andy (nevercontent)

A little typo, Andy is buynholdisdead :wink:

Neil

Also he isn’t anywhere close as bright as nevercontent :slight_smile:

Andy

6

I responded making clear that wasn’t implying that it IS a big deal. I have no way of telling if it’s a big deal or not. I was just saying that they have to SAY that it’s no big deal, whether it is no not! That’s a different issue. It just means you can’t bank on what management is currently saying.

I think the problem I have with this logic is that first you have to believe that all management will lie to you. Second you have to believe that defining new commands for your switch is some how a huge undertaking.

First, Anet has the best management in the communications and networking field. There isn’t another management team that can even touch them. That is why Cisco is so afraid of them. It’s not only their technology which is very impressive.

Second, have you ever used cli commands? Think of it as doing Dos commands. I used to have to memorize commands from each different vendor in order to program any piece of equipment I worked on. The commands from vendor to vendor were closely related but different enough to make it a headache when working across different pieces of equipment. So what has changed? I rarely use ClI commands any longer. Most of my work is done in a gui interface. The gui interface, when you click on a button, executes the CLI command behind the interface. I will bet anyone, right now, that ANET is working on a fix for this and that it will not affect their gui interface in the least. Most technicians will not even know that the commands have changed.

Finally, their moat has nothing to do with their hardware. The hardware is now a commodity with many companies going to white boxes and loading their own software. FB started this trend but the one thing FB has not done is quit buying Anet’s product. Why? Because of their EOS software. Does anyone realize that Anet has many more software engineers than they do hardware engineers? That they just recently started selling their software as a monthly subscription. That this software will allow you to provision many ports across your network with one single command rather that having to provision each single port individually?

I strongly doubt that they and Cisco would each be paying millions of dollars a quarter in legal fees if it was just a question of changing some wording of commands, and it was not going to matter much if they won or lost. I figure that this may be closer to an existential threat IF they were to totally lose the suit. That’s a big IF, of course, but I wouldn’t totally discount it considering that they lost this first, very preliminary decision.

Well Saul this is where we just have to disagree. I think Cisco would sue Anet just to break them. Why not? They are a strong competitor which obviously Cisco is scared of. They have never sued another competitor for using their commands. Also to call this an existential threat is way over the top. You may think it would be hard for them to change the commands but in reality it would be very simple for a company like ANET to do this. In the short term this might hurt Anet’s stock but in the longer term I do not think it will have much of an effect.

Andy

22 Likes
7

I am not wading into this to say one way or another is correct, ANET just came on my radar and I haven’t done the diligence to hold any conviction of it.

BUT I do work with this type of hardware and software a lot, and it really isn’t that big of a deal learning a new vendors products. It kinda goes two ways, you got the guys/gals who love learning and coding and get excited about CLI, and you got others who are super comfortable with what they know and don’t want to change (they already love/like a way and don’t want to switch, which is great). No knock on either, btw, I have fallen into both ‘camps’ multiple times in my career, personally I bend towards humility and not pride in these decisions.

For some their is a GUI, for others a CLI. This ‘debate’ has gone on since the dawn of computers. My point though, is it won’t be an issue to implement the products. Management buys something, technicians implement it, and rarely the two shall meet. If they are Cisco people, and their company no longer uses Cisco, they can move on easily. These are highly employable people. Most of the technicians I know only work on 1 vendor for 5-10 years at a time, or maybe for their whole career. Cisco did a great job with their CCNA certificates, Juniper is doing the same, and Arista can as well. It isn’t just in network hardware/software either. I would dare say every IT company/sector is like this. People fell on the Microsoft vs. Apple sword for decades (I like Linux). If you are interested and good with Splunk you will be employed for decades, and never have to learn another vendors products/language/CLI/GUI. Same thing with security vendors like McAfee, once you get hooked (trained and experienced) on their product line you are employable. Heck, the vendors themselves will hire you away if they find you competent (usually to fill in the gaps for the people that left when companies switched vendors).

All that to say, I wouldn’t put a lot of weight on that assumption with Arista. Now the legal issues, trust, market share, management, growth, financials, 1YPEG, etc. builds/destroys conviction in a company.

Robert
no position in ANET, but looking

6 Likes