Try a WHOIS blacklist for domain generation

[turtlesoupy]

[AndreiIgna]

a good idea

[JonLaliberte]

Any idea where to get such a list?

[turtlesoupy]

@JonLaliberte I didn't actually find anything good -- it is easy enough to do dynamic lookups but it gets expensive since the code is designed with a master list in mind. Let me know if you find anything!

[hartshorne]

ICANN publishes most zone files here: https://czds.icann.org/ — the .com zone has over 100 million domains in it. I'm not sure that training on domain names will get great results. There is a lot of junk in there.

I generated some words doing something like:

from title_maker_pro.word_generator import WordGenerator
word_generator = WordGenerator(
  device="cpu",
  forward_model_path="<somepath1>",
  inverse_model_path="<somepath2>",
  blacklist_path="<blacklist>",
  quantize=False,
)

# a word from scratch:
for id in range(30):
  print(word_generator.generate_word())

From this set, over half of the words generated were available to register:

Name	Status
bakilaka.com	available
biopsychosomatic.com	available
boepland.com	available
bunchkine.com	available
carrumbas.com	available
chicanto.com	available
dozohedron.com	available
endomautics.com	available
hydropathymetry.com	available
miserically.com	available
plautay.com	available
predaconal.com	available
prigue.com	available
sarappilla.com	available
sockmeal.com	available
unadipated.com	available
wickety-wick.com	available
adraven.com	registered
affella.com	registered
backblast.com	registered
backround.com	registered
diarism.com	registered
hoive.com	registered
jointness.com	registered
laceway.com	registered
midface.com	registered
presentant.com	registered
supermicrobe.com	registered
tansi.com	registered
windbow.com	registered

Not bad!