Well ArcaOS 5.0 is probably the last version for me.

[Dariusz Piatkowski]

Roderick, everyone...

Again I am looking for numbers to estimate how big our community roughly but it will always be a gamble to a large extend.
...
The 500 users was based on the IP numbers from Netlabs. But how accurate that is ? But I never stated we have 500 private users left. Read the other postings for the rest of the details...

So why don't we set up a honey-pot of sorts and use a different approach to uniquely identify the hardware that's running OS/2?

My understanding is that ANPM does exactly that. They use dmidecode to pull an OEM string that identifies the hardware, my understanding is that this is the UUID field value (not 100% sure, but someone could verify).

So here is what that looks like on my machine:

Code: [Select]

Handle 0x0001, DMI type 1, 27 bytes
System Information
        Manufacturer: MSI
        Product Name: MS-7640
        Version: 1.0
        Serial Number: To Be Filled By O.E.M.
        UUID: 00000000-0000-0000-0000-6C626D50A7A8
        Wake-up Type: Power Switch
        SKU Number: To Be Filled By O.E.M.
        Family: To Be Filled By O.E.M.

Take a look at the full dmidecode dump (see attachment). What would simplify this is having a front-end web page that spells out how the end-user should do this, and we're talking literally a simple 1-5 step process, or simply accepts the UUID value that the user extracts out of the dmidecode dump.

Heck, dmidecode has an option to literally pull just the UUID, that being: "dmidecode --string system-uuid".

[Neil Waldhauer]

Counting the number of users more or less useless. IP is better than UUID. UUID counts user's systems, not users. For instance, I prepare systems for customers by updating from NetLabs. I'm just one guy, but you will count every system I attempt to prepare.

It's better to count contributing users. It's better still to make it easier to be a contributing user.

Starting out by trying to count users to demonstrate that someone is getting NetLabs without contributing is not useful; it is a sign of mental illness. Maybe take a vacation or something if you are counting users for the purpose of finding out how many "freeloaders" there are.

If the goal is permanent full time employment for BitWise, think about selling subscriptions and publicly showing the results. I think we agreed that 300 subscribers at 10 Euro would finance them indefinitely. (I think 200 at 10 Euro and 10 at 100 Euro is more reasonable) This is a much better thing to count than IP or UUID. It would be better still if someone at BitWise was involved and running this.

[ivan]

As it stands this whole question of the number of users should be considered as a marketing exercise.  That should provide a greater incentive for those porting programs.

The last time I looked at the Netlabs RPM repository (using one of my Linux boxes) the bulk of the items there were things I will never be interested in (programming languages and libraries) rather than the things I am interested in (example: a full QT5 runtime environment that fits in one directory like the QT4 runtime and 'just works' - the QT4 was a superb effort, the QT5 appears to have fallen apart by concentrating on parts of the runtime rather than the whole [sorry if that upsets anyone but that is the way I see it]). 

As I said, consider the whole thing as a marketing exercise, find out what 'we the users' really want, not what we have to put up with.