But your suggestions are not good. It's just "stick words on the front or end.
The simplest thing you can improve is include variants of the word. If I say "kayak", then kayaks and kayaking are closer to my intent than "wikikayak". Getting this right would be your secret sauce.
Obviously, non .com domains would make this much more valuable.
add a filter of "only show available domains" would be very helpful.
maybe have the option to do prefix and suffix? Probably only useful with AI or some algorithm that makes them related. MyKayakHome isn't bad.
On the quirky tab, you don't display the domain names, you show "sug+". I assume this is supposed to be added to the stem, but it's not consistent. "a+" guard gives "aguard", but "sug+" guard gives "suguard", where is the extra G?
Either way, please just show the domain name.
If I select the text for the name of a domain, you should not pop up the domain info box.
One the ai tab, I got "An error occurred in the Server Components render. The specific message is omitted in production builds to avoid leaking sensitive details. A digest property is included on this error instance which may provide additional details about the nature of the error."
On the main page, the search button is always disabled, but hitting enter works (firefox/mac). hmm not reproducible after reload.
On firefox mac, sometimes the search button stays disabled. I don't have a way to reproduce, but it's shown up several times. I just reload the page. You can still hit enter and search, though.
It looks like you are just adding the same list of words as a prefix, in the same order, no matter what the base word is. That seems too simplistic to me.
Some words are more likely to occur in conjunction with specific words. Kayak and water, for example is probably more likely than kayak and tiger.
You might try something like a collocations data set like https://www.freecollocation.com/ (I just searched for dataset of English word collocations)
Any autocomplete that words more than one word at a time has to do this. You might check that out.
I’ve tried creating similar solutions and feel LLMs still lack accurate control over (or understanding of) length, stress / accents, and phonetics for consistent name generation. For usernames for example I’ve yet to create a generator that uses LLMs that beats simple Markov chains. Maybe because results are subjective it makes rating / training a lot harder? I like the site and your approach though and great job on lookup speed! If anyone has any tricks or suggestions I'd love to hear them.
I thought the same but noticed it doesn’t even check all the domains unless you scroll down the list and it lazy loads each on as they appear in the viewport.
OP. Sorry to be harsh but this is awful UX in terms of reviewing search results. Every thing I search said 0 available so I searched again and again. Then I scrolled down and noticed they were being checked as I scrolled and the number of available domains matching actually become >0. Not how i want to see the results.
FWIW, I’ve had similar complaints about LDS and a few lines if javascript will clean up their search result page (exclude .blog and unavailable). Not able to share rn but look at their dom and im sure you could have done this too.
>OP. Sorry to be harsh but this is awful UX in terms of reviewing search results
No problem, it can be improved by using the Verisign .com domain list that is updated daily. I just wanted to try something that I could ship in a couple of days after learning about RDAP and that it allowed client side requests.
If there's interested in the service, I'll improve it.
Thank you, Chris! I love exploring options using the suffixes/prefixes [1] lists from wikipedia and InstantDomainSearch, and this saves me a lot of time. Hope you got some value from it and found some interesting names!
>Could be worth adding some kind of 'only show available domains' option
Yes, that will require a rework, but I'm willing to do it if this gets any traction.
This is useful especially with the weird .blog insertion on LeanDomainSearch. But I've always wondered how LeanDomainSearch gets results so quickly? The results are almost instant for thousands of domains.
They likely use a batch API or maintain a local cache of recently checked domains, combined with prefix/suffix filtering against a pre-computed availability database rather than doing individual RDAP/WHOIS lookups for every possible combination.
I didn't know RDAP lookups was a thing, but now i've learned something by looking at your source. Thanks OP
https://www.verisign.com/en_US/domain-names/registration-dat...
You're welcome! I did not know either until very recently, that's why I decided to give this side project a try.
Some feedback:
Very cool! I will use this.
But your suggestions are not good. It's just "stick words on the front or end.
The simplest thing you can improve is include variants of the word. If I say "kayak", then kayaks and kayaking are closer to my intent than "wikikayak". Getting this right would be your secret sauce.
Obviously, non .com domains would make this much more valuable.
add a filter of "only show available domains" would be very helpful.
maybe have the option to do prefix and suffix? Probably only useful with AI or some algorithm that makes them related. MyKayakHome isn't bad.
On the quirky tab, you don't display the domain names, you show "sug+". I assume this is supposed to be added to the stem, but it's not consistent. "a+" guard gives "aguard", but "sug+" guard gives "suguard", where is the extra G?
Either way, please just show the domain name.
If I select the text for the name of a domain, you should not pop up the domain info box.
One the ai tab, I got "An error occurred in the Server Components render. The specific message is omitted in production builds to avoid leaking sensitive details. A digest property is included on this error instance which may provide additional details about the nature of the error."
On the main page, the search button is always disabled, but hitting enter works (firefox/mac). hmm not reproducible after reload.
Hey I've made a similar tool that shows only available domains and I've hooked it up to gemini to generate creative domain name, check it out!
https://www.domize.io/
The error ""An error occurred in the Server Components render. " comes from putting the description in the word input field.
On firefox mac, sometimes the search button stays disabled. I don't have a way to reproduce, but it's shown up several times. I just reload the page. You can still hit enter and search, though.
It looks like you are just adding the same list of words as a prefix, in the same order, no matter what the base word is. That seems too simplistic to me.
Some words are more likely to occur in conjunction with specific words. Kayak and water, for example is probably more likely than kayak and tiger.
You might try something like a collocations data set like https://www.freecollocation.com/ (I just searched for dataset of English word collocations)
Any autocomplete that words more than one word at a time has to do this. You might check that out.
Thank you so much for your time writing all this feedback, Tom! I did not know about "collocations", and I'll certainly check that out.
I’ve tried creating similar solutions and feel LLMs still lack accurate control over (or understanding of) length, stress / accents, and phonetics for consistent name generation. For usernames for example I’ve yet to create a generator that uses LLMs that beats simple Markov chains. Maybe because results are subjective it makes rating / training a lot harder? I like the site and your approach though and great job on lookup speed! If anyone has any tricks or suggestions I'd love to hear them.
Thank you! Maybe keep generating using the Markov Chains and use a LLM to evaluate the results?
I have 1 suggestion, as domain order is not important, can you move available domains to the top as soon as you found them available.
thanks
I thought the same but noticed it doesn’t even check all the domains unless you scroll down the list and it lazy loads each on as they appear in the viewport.
OP. Sorry to be harsh but this is awful UX in terms of reviewing search results. Every thing I search said 0 available so I searched again and again. Then I scrolled down and noticed they were being checked as I scrolled and the number of available domains matching actually become >0. Not how i want to see the results.
FWIW, I’ve had similar complaints about LDS and a few lines if javascript will clean up their search result page (exclude .blog and unavailable). Not able to share rn but look at their dom and im sure you could have done this too.
>OP. Sorry to be harsh but this is awful UX in terms of reviewing search results
No problem, it can be improved by using the Verisign .com domain list that is updated daily. I just wanted to try something that I could ship in a couple of days after learning about RDAP and that it allowed client side requests.
If there's interested in the service, I'll improve it.
Love this, thanks for building it, way faster way to explore options than what I'm used to!
Could be worth adding some kind of 'only show available domains' option
(edit: realizing this may require a rework of how the underlying availability checking process works, but could be worthwhile)
Thank you, Chris! I love exploring options using the suffixes/prefixes [1] lists from wikipedia and InstantDomainSearch, and this saves me a lot of time. Hope you got some value from it and found some interesting names!
>Could be worth adding some kind of 'only show available domains' option
Yes, that will require a rework, but I'm willing to do it if this gets any traction.
[1]: https://en.wiktionary.org/wiki/Category:English_prefixes
This is useful especially with the weird .blog insertion on LeanDomainSearch. But I've always wondered how LeanDomainSearch gets results so quickly? The results are almost instant for thousands of domains.
There's an old blog post[1] by the founder of LeanDomainSearch on this.
[1]: https://mattmazur.com/2018/05/18/extracting-a-list-of-all-re...
They likely use a batch API or maintain a local cache of recently checked domains, combined with prefix/suffix filtering against a pre-computed availability database rather than doing individual RDAP/WHOIS lookups for every possible combination.
>But I've always wondered how LeanDomainSearch gets results so quickly?
You can sign up with Verisign and they let you download a text file with all the registered .com domains, updated daily.
I found a bug: if you include a . in the domain your looking for (ie a.b), all domains will show as available, even though they aren't.
Will fix, thank you!
Small suggestion, in the simple tab you could put the <term>.com of the term searched as the first result. Cool project.
Thanks, Mickey, for trying and for the suggestion!
yeah, if the base foo.com domain is available, that's the most important thing. I don't see that.
Great idea overall. The results in the AI tab are not very creative though, just variations of the same creative idea.
Thanks for trying! The AI tab is my least favorite as well, will try improving the prompt and changing to a better model.
Loved it, thanks for sharing it!
Thank you! Hope you found a great name!
Love it--feel like it helps builders fight back against domain squatters.
ai tab Generate button is disabled for me, regardless of input
What browser are you using, please?