Advantages, disadvantages of installation

[RMerlin]

In my mostly 386 code base experience I never encountered a case where something was broken in Asuswrt, but was working in then available Asuswrt-Merlin release. I would like to have examples of such cases so I can test and see.

The GT-AXE16000 webui, specifically the Wireless -> Professional page as of right now. The current latest stock firmware is broken, with various settings not showing up for the correct radio. The fix should be available in the next stock release.

I believe you are correct. Asuswrt comes with more and more Asuswrt-Merlin features included. My concern is this cooperation will turn Asuswrt-Merlin firmware into more niche option for mostly advanced VPN configurations and custom scripts.

Keep in mind that even in the early days, my focus was never about implementing tons of new features. I was specifically recommending people to look at DD-WRT back then (as DD-WRT was still a valid option at the time) for anyone looking for tons of new features. It was clearly stated in the documentation. That hasn't changed. The core of the project has always been about tweaks, QoL enhancements (like the vastly more useful Wireless Log page, or editable DHCP reservations), with just a few occasional new development that remain exclusive, like VPN Director or an addon API. A one-man project of the scope of this firmware simply doesn't have the resource to both do maintaining, adding new models, AND developing new features. The 388 merge is expected to take a good 3-4 weeks of my time, and that's just a rough guesstimate as I haven't seen the code yet, but it's what it took me for the 386, and before that the 382 merges.

 

[Akore]

The GT-AXE16000 webui, specifically the Wireless -> Professional page as of right now. The current latest stock firmware is broken, with various settings not showing up for the correct radio. The fix should be available in the next stock release.


Keep in mind that even in the early days, my focus was never about implementing tons of new features. I was specifically recommending people to look at DD-WRT back then (as DD-WRT was still a valid option at the time) for anyone looking for tons of new features. It was clearly stated in the documentation. That hasn't changed. The core of the project has always been about tweaks, QoL enhancements (like the vastly more useful Wireless Log page, or editable DHCP reservations), with just a few occasional new development that remain exclusive, like VPN Director or an addon API. A one-man project of the scope of this firmware simply doesn't have the resource to both do maintaining, adding new models, AND developing new features. The 388 merge is expected to take a good 3-4 weeks of my time, and that's just a rough guesstimate as I haven't seen the code yet, but it's what it took me for the 386, and before that the 382 merges.

So if you retire there will be no more updates or will you pass the torch on to someone else?

 

[RMerlin]

So if you retire there will be no more updates or will you pass the torch on to someone else?

All my code is on Github. Anyone can already fork the code if they wished to.

 

[Yota]

It probably did increase, but I have no easy way to get accurate numbers.

I coded a simple method for you to get the exact number.

firmware update: add unique identifier for statistics by JackMerlin · Pull Request #821 · RMerl/asuswrt-merlin.ng

Add merlin_fwupdate_uuid mvram variable to persist a UUID, UUID is randomly generated, not bound to hardware, and will be reset after factory reset. Statistics sent to the server are sent through t...

github.com

You just need to look at the user agent log in your server admin page, export it to excel to get statistics simply.

This is just a proof of concept and there is still a lot of work to be done as it is uncertain whether this is in compliance with European privacy protection laws.

 

[RMerlin]

I coded a simple method for you to get the exact number.

I don't need a user-unique ID, I just need to check the number of queries sent to the update server within 24 hours. Unless people were to reboot their routers multiple times per day, that alone would provide a fairly accurate report.

 

[Yota]

I don't need a user-unique ID, I just need to check the number of queries sent to the update server within 24 hours. Unless people were to reboot their routers multiple times per day, that alone would provide a fairly accurate report.

Well, just including the model and firmware in the user agent would be enough. This will have no privacy risk and can be used for statistics.

Not sure what your CDN provides to show you user-agent based access information, but you should always be able to export logs from the CDN and make your own graphs.

 

[Tech9]

A one-man project of the scope of this firmware simply doesn't have the resource to both do maintaining, adding new models, AND developing new features.

For one-man project Asuswrt-Merlin is a huge achievement indeed. I'm under impression Asuswrt is cannibalizing Asuswrt-Merlin. Except more advanced configurations in VPN Director and addon API, almost everything else is made available in 388 code with the same description and at the same places in GUI pages. More average users just have less reasons to use your firmware. This is your work and I believe Asus owns you early retirement. And an Audi RSQ8 at least, for Christmas 2022 after 388 official release.

 

[RMerlin]

For one-man project Asuswrt-Merlin is a huge achievement indeed. I'm under impression Asuswrt is cannibalizing Asuswrt-Merlin.

I wouldn't say "cannibalizing" because this is done with my support. About two years ago, Asus and I discussed a list of potential features that would make sense for them to integrate into the stock firmware. One of the rationale at the time was that by being included in the upstream code, it would make things a bit easier for me to merge (with the code being a bit less different). So, this is done with my full knowledge and support.

Yes, it makes Asuswrt-Merlin a bit less unique and advanced. But my goal isn't to sell a competitive product here - I am not in competition with Asus. Therefore, I'm fine with it.

More average users just have less reasons to use your firmware.

Personally, I consider that overall the biggest unique feature of Asuswrt-Merlin, even if not as in-your-face as something like DNSFilter, remains customized scripts and configs. That alone makes Asuswrt-Merlin a major step forward ahead of the stock firmware for many advanced users.

I'm also fairly sure that things like my OpenVPN implementation will generally remain ahead of the stock firmware implementation, as I will keep focusing on advanced users, while Asus need to make it as user-friendly as possible for the average user. Different design priorities.

More average users just have less reasons to use your firmware.

And I'm ok with that. If Asuswrt-Merlin were a product that I am selling, or relying on revenues generated by it to earn a living, the situation would be different.

This is your work and I believe Asus owns you early retirement. And an Audi RSQ8 at least, for Christmas 2022 after 388 official release.

Maybe Asus, as a company, benefit more from me than I do from them, but that's just the way it is with manufacturers benefiting from open source projects. Otherwise, I would say that the developers of dnsmasq or busybox would be owed a lot more by all router manufacturers. Where would be NAS manufacturers if Samba didn't exist?

 

[RMerlin]

Not sure what your CDN provides to show you user-agent based access information, but you should always be able to export logs from the CDN and make your own graphs.

Cloudflare requires a $200/month Enterprise account to get log data.

 

[Tech9]

but that's just the way it is

You guys work for free for someone who runs a business for profit. The way has to be changed.

 

[Yota]

Cloudflare requires a $200/month Enterprise account to get log data.

My fault, I didn't realize they don't provide logs for free, I thought it was a basic service.


Ok, but I just googled and there are 3rd party services out there that are offering logging services for free, of course this requires embedding 3rd party stats code in the website, (like this: https://logflare.app ) I don't know if this is possible.

Otherwise, having the router actively send a special request to the server for the server to log is beyond my control.

Anyway, sorry for not helping.

 

[RMerlin]

You guys work for free for someone who runs a business for profit. The way has to be changed.

We (open source developers) are doing it as a hobby. Nobody is forcing us to do it, we do it for our own entertainment or for other personal reasons, and if someone can benefit from our work, then good for them. It doesn't take away anything that we are currently getting out of it.

I have been working on freeware/opensource projects for most of the past 30 years, with a period of about 5-10 years where I wasn't actively involved in any project because work was my priority during that time.

 

[Tech9]

I get the philosophy, but I still find it unfair. Asuswrt-Merlin in particular isn't a small project. It takes significant amount of your time, requires hardware investments (your PC didn't come for free), requires connections to keep alive and on top you provide support dealing with questions when and why. Apart from big manufacturers taking advantage of your (including other developers) work, some small players openly talk about running for profit businesses around Asuswrt-Merlin and custom scripts right here on SNB Forums. Did they set a monthly donation for you guys or you get Likes mostly? A global player like Asus can easily afford a donation big enough for a house in Westmount for you. You can practice your hobbies from there, no?

 

[Yota]

I get the philosophy, but I still find it unfair. Asuswrt-Merlin in particular isn't a small project. It takes significant amount of your time, requires hardware investments (your PC didn't come for free), requires connections to keep alive and on top you provide support dealing with questions when and why. Apart from big manufacturers taking advantage of your (including other developers) work, some small players openly talk about running for profit businesses around Asuswrt-Merlin and custom scripts right here on SNB Forums. Did they set a monthly donation for you guys or you get Likes mostly? A global player like Asus can easily afford a donation big enough for a house in Westmount for you. You can practice your hobbies from there, no?

I think there is too much investment out there on this project, Eric has invested a decade in this project, PC, cables for debugging, websites, and even the CDN we discussed above.

Tim's hosting of the SNB forums is also not a small expense, the XenForo forum is not free, not to mention the domain name and server expenses.

There are also many technical support members, script developers, and fork maintainers here.

These people used their time and money to make this project possible, benefiting hundreds of thousands of users.

So yeah, I agree, it's not fair.

But that's why open source projects are so great, not just open for anyone to participate, but more freedom, freedom to participate in this project. there are no rules out there saying what you can't do and no rules saying what you should do, if one day Eric or some developer gets tired and decides to stop his work, no one should complain because it's his choice.

Developers who invest in open source projects don't measure their willingness to participate with money, they have their own metrics, maybe the number of stars on GitHub, maybe the number of replies from the thread, maybe just for the Like or fun.

In any case, their choices have contributed to free and open source software, creating a community where more like-minded people can join. For the Merlin project, although the main developer is Eric, this spirit of free participation has been integrated into this community. We do have many people actively replying to threads on the SNB forum, providing free technical support, developing add-ons, and helping other users.

I would say that Eric influenced all of this not only with his work, but more with his ethos of open source software.

As beneficiaries, I think we should all give back to this project if possible, perhaps to give developers an affirmation and gratitude, but more importantly to help the project in a way that suits you. And never, never break the open source flames, because those illegal forks, attempts to port to unsupported models are the force against this project.

Speaking of Asus, should Asus start sending checks for Eric? maybe, but not should be, they will decide what kind of giving back they can do, but they also have restricted areas, they should always respect and abide by Eric's wishes, and for them, obeying the license set by Eric will be the foundation of everything.

Ok, here are my two cents as an Asuswrt-Merlin beneficiary over the years.

 

[thelonelycoder]

some small players openly talk about running for profit businesses around Asuswrt-Merlin and custom scripts right here on SNB Forums

Do you have links to these? amtm is open source and will only support scripts that are too.

 

[Martinski]

...
Developers who invest in open source projects don't measure their willingness to participate with money, they have their own metrics, maybe the number of stars on GitHub, maybe the number of replies from the thread, maybe just for the Like or fun.

Fully agree. The Open Source software community is not in it for financial gain, and the notion of "closed sources & software for profit" is considered an anathema to their philosophical tenets. Yes, their philosophy may be counterintuitive in a capitalist economy/society, but that's the way the Open Source community sees it. Whether or not other people take advantage of their work & contributions to Open Source is ultimately not relevant or important to their principles.