They are very hit or miss. On balance, I don't bother with them, unless there's something interesting looking on their daily free book page, or it's one of these bundles where they come out to $.75 apiece.

Some of the ones I've gotten don't look like they ever got near an editor, or even a spell-checker.

If I'm going to spend money and time, I'd rather go to Manning and have some confidence that it'll be quality.

When they contacted me to write for them I did a bit of research I found enough horror stories to not follow up. In fact it just confirmed the feeling I got as a reader.

Go with Manning, PragProg, NoStarch or even (with some caution) O'Reilly if you want quality content.

As someone who writes for Packt (Posting under a throwaway for obvious reasons), I think you made the right decision. The first title I was proud of, had good editorial control and decent technical review. The second title had an awful premise that was insisted upon and stunningly poor editorial control. Disappointingly I now feel like I have to apologise for something that carries my name.

My next title will be self-published onto Github on a 'freemium' model. It'll allow me to develop the content in the open, and if people are so disposed, they can pay for an edition that carries additional explanatory text. Either way, I won't be writing for Packt again.

When I worked at ThoughtWorks, I also got approached by them, and told my sponsor and colleague about it. Then the horror stories came, and mostly about the time pressure they put on you and the miserable pay-off. Most of the time you will be rewarded with give-away copies...

I haven't read any of these, so I can only answer indirectly. All the other Packt books I've seen have been mostly crap. Badly written, poorly edited, not really worth your time.

I know some authors it sounds similarly poor from their perspective as well.

I have reviewed books, and the input I gave all been to a deaf man's ear. comments are collected in a word document and it feels the author never got any of them as even the code examples still contained issues I pointed out during the writing process.

While some of the authors really put effort into it, I do not believe the quality of the books is of a very high standard. I'd rather spend my money elsewhere...

My experience with Packt has been wholly negative.

To give an example of what to expect, my experience with one particular book on Scientific Python: * most code examples wouldn't run without significant modification, mainly due to missing lines and multiple typos * some of the text was cryptic to the point of absurdity * topic coverage was unstructured, patchy and made no coherent sense overall * the website for the book was broken and I couldn't find any way to feed my corrections back to the author

Clearly Packt put no effort into quality assurance or editing. I would suggest you would be better off waiting for something from a more reliable publisher: such poorly edited books will just waste your time.

I got offered to write a packt book about a subject I don't know shit about, through LinkedIn, by one of their editors, just because I did an internship in a loosely related field. I feel like they're desperate to churn out books and literally take anyone with anything that could be loosely interpreted as experience in a related field as authors

Just a random thought from your comment, but it feels like stuff like this is one reason quality physical books will not be going away. I mean, I doubt they're making print editions from all this crap...?

I bought their book on "Deep Learning with Keras "and found it very useful. Lots of hands on examples, little to no theory. Since there is a dearth of books on Keras, this served my needs perfectly.

I skimmed 'Building machine learning systems with Python' once before. It looked worth having.

People aren't desperately throwing themselves to buy them for their full price it seems, because Packt is always there with humble programming books of any sort.