What is a nanode?

Yep, the 5$ / 1GB Linode plan got a cool new name!

In short, 'Nanode' is a portmanteau of 'Nano' and 'Linode'. Nano because it's small, and Linode cause it's hosted with us!

Alright, so it is the same plan with a cute new name. That's cool - but honestly, it feels pretty sleazy to rearrange the UX of a page that is directly tied to how much a customer spends, without warning; particularly when the motivation behind the change is clearly to drive users to spend more. I've already accidentally deployed a 2048 once because it's now in the spot that the "Nanode" occupied before.

You could make the argument that I should look more closely - and really you wouldn't be wrong - but rearranging the UX to try to grab a few extra dollars from people just feels like a tactic I might expect from a lesser provider. I would probably shrug it off in that context, but it seems really out of character with the overall quality of Linode's product.

Just bumping because this is still an obviously dark UX pattern and it makes Linode look really seedy.

rearranging the UX to try to grab a few extra dollars from people

That wasn't the intention, but I can see how that's what it looks like. We felt this was the best way to separate out the plans in the existing Manager design, which didn't have different sections for different plan "types" in the past. BUT - I totally get where you're coming from. We're working on a complete redesign of the Manager which accommodates for different plan types from the start, and we'll keep this feedback in mind.

That wasn't the intention, but I can see how that's what it looks like.

It's easy to make that assumption, as a Linode 1024 only became a "Nanode" on the pages where you purchase things - on the main pricing page, for example, it's still listed with the other plans:

https://www.linode.com/pricing

Still a standard plan here, too:

https://linode.com/docs/platform/billing-and-payments/#standard-plans

We're working on a complete redesign of the Manager which accommodates for different plan types from the start

I don't see what issue is being addressed by creating a separate plan "type" for the Linode 1024. If this isn't a change that's designed to fool users into buying a larger plan, then what is the intended benefit of this change, for customers? Because having only one plan (and I'll venture to say it's a commonly used plan, though obviously I don't have access to those numbers) separated from the rest arbitrarily (we've established there's no technical difference between this plan and others, indeed in terms of resources it's very similar to the old $10 plan) doesn't really seem like a change that benefits customers - it seems inconvenient at best, and deceitful at worst.

on the main pricing page, for example, it's still listed with the other plans:

Good catch, our pricing page/doc might need updating – we’ll look into this.

I don’t see what issue is being addressed by creating a separate plan “type” for the Linode 1024. If this isn’t a change that’s designed to fool users into buying a larger plan, then what is the intended benefit of this change, for customers?

We started experimenting with different plan types with the introduction of High Memory plans – we’re going to keep doing that with different plans, and this change is just part of that. Part of this process is experimenting with expectations about what each of our plans is meant for – for example, the use case of a 1GB Linode isn’t the same as the use case of a 200GB Linode. We’re actively thinking about our plans, so in the future, you could see further differentiation in plan types.

it seems inconvenient at best, and deceitful at worst.

It's entirely possible the new Manager will undo this change. Or we'll differentiate our plans in some other way. Or something we haven't thought of yet. This is an active process, and nothing is set in stone, so feedback on this from the Community is appreciated.

1GB Linode isn’t the same as the use case of a 200GB Linode

Sure, but that's a pretty extreme comparison. I doubt there's much confusion on the part of customers as to whether they need a 1GB plan or a 200GB one. However I'd say the use case of a 1GB is pretty similar to that of a 2GB Linode - similar enough that there's no reason for the 1GB to be hidden away. I doubt that it's a common issue for customers to deploy a 1GB when they meant to deploy a 2GB (though the reverse is now true, since the 2GB occupies the spot that the 1GB once did) - which gets back to the question of what problem this is actually solving for customers.

We started experimenting with different plan types with the introduction of High Memory plans

Again, not really an apt comparison - the high memory plans were new plans, geared toward memory-intensive tasks. The nanode is just a reclassification of a single existing plan based on, as far as I can tell, nothing other than its reduced price point.

Part of this process is experimenting with expectations

I didn't really sign up to be a part of this experiment - if you want to force weird UX changes into your beta manager that's cool, it's your beta, and I'll register my feedback appropriately. But this change was forced into the existing manager as well - without warning - breaking the expected UX. Obviously Linode's customers come from all sorts of backgrounds, but I think it's safe to assume (as the pricing page and much of your other marketing materials do) that the average Linode customer can understand that 1GB is less than 2GB, without a cute renaming that hides the cheapest plan at the bottom of the page.

Fair enough. Thanks again for the feedback – and we need to hear from everyone on stuff like this, so I invite anyone else with comments on this change to sound off here.

We have lots of good stuff in the works – stay tuned.

Thanks for your infos… but can we be sure that there is no difference (such as hardware types) between nanode and standard ones?

Nanodes are hosted on the same machines as our other plans, so there's no difference in equipment, only allocation of resources.

You can check what hardware your Linode is operating on with the following command: sudo lshw

Hi.

I was also a little bit confused about the name change, so I googled for "nanode linode" and this thread was the first result.

The reason I don't like the new name is that it suggests that the hardware/performance may be different.

In particular, some cloud providers have servers with shared CPU (i.e. less than one CPU). For example, in AWS they are called T2 instances, in GCE they are called g1-small or f1-micro, and the whole Amazon Lightsail service is made of shared CPU instances.

Those instances are usually "good enough" to host a small traffic web server (while still being more expensive than Linode :-), but not for building Debian packages from source, which is one of my use cases.

So, am I right to think, from the comments above, that "Nanodes"
are not shared CPU instances or anything like that?

Thanks.

Hey sanvila!

So, am I right to think, from the comments above, that "Nanodes" are not shared CPU instances or anything like that?

CPUs for all Linodes – including Nanodes – are a shared resource. You share CPU time with other guests on the physical host.

A few answers up, I talked about how we might differentiate plan types:

We started experimenting with different plan types with the introduction of High Memory plans – we’re going to keep doing that with different plans, and this change is just part of that. Part of this process is experimenting with expectations about what each of our plans is meant for – for example, the use case of a 1GB Linode isn’t the same as the use case of a 200GB Linode. We’re actively thinking about our plans, so in the future, you could see further differentiation in plan types.

Nanode is a new plan family, and we have additional new plans families in the works. Our current thinking is that, in the future, Nanodes will be for "low CPU" use cases, while additional families will be introduced.

As it currently stands, though, a Nanode doesn't have its CPU artificially capped differently than other Linode plans, and I think that's what you mean here. That'll change in the future as our product line evolves.

I hope that helps answer your question!

As it currently stands, though, a Nanode doesn't have its CPU artificially capped differently than other Linode plans, and I think that's what you mean here. That'll change in the future as our product line evolves.

So if I purchase a Nanode right now, while it still has the same CPU allocation as a 2GB plan, can I assume that in the future a "free upgrade" might move me to plan that is artificially capped?

Also, why not list the Nanode first on the "Add a Linode" page, since that's where's it's listed on the pricing page: https://www.linode.com/pricing#all

(Never mind that it's still listed as a standard plan there, which seems to defeat the purpose of this whole psychology experiment.)

I second what smallclone is saying and how they are saying it. Inconvenient at best and deceitful at worst. I also wondered why the $5 box had been moved to the bottom. My experience with organizations like linode is that the center cannot hold and you must gradually go downhill. So when you have a past history of being rather lovely with a straight and fair passthrough that does not play favorites or muddy up your editorial choices with technical choices, this in a way is "noncommercial" because you are eschewing the chance to maximize your commercialization. Then if you switch and do this more, you're going downhill. You've given the money and revenue people a seat at the table, more than they had in the past. "Dark UX" is a fair characterization. Please do not do more of this. Users can smell the difference.

That is my response to jackley saying "we need to hear from everyone on stuff like this, so I invite anyone else with comments on this change to sound off here."

Hi folks! We've made another change to the "Add a Linode" page. "Nanode instances" are now positioned at the top, before our other instance types. Thanks for the feedback!

Also, @smallclone, I realized I missed this point – sorry!

So if I purchase a Nanode right now, while it still has the same CPU allocation as a 2GB plan, can I assume that in the future a "free upgrade" might move me to plan that is artificially capped?

Nothing is yet set in stone on this.

kewl