5 ms·

As many others have said, books are the best things I've spent money on, but let me say more: Do books sometimes say things that are obvious? Yes. For example,

As many others have said, books are the best things I've spent money on, but let me say more: Do books sometimes say things that are obvious? Yes. For example, when I first read Martin Fowler's book Refactoring, I had been renaming variables and moving methods from one class to another for years. But he gave a new framework for thinking about something that obvious. I've found the idea of separating out my coding flow between adding functionality and improving code really helpful. Can you get the material for free? Almost always, yes. But it takes time to find the right material, and our time is valuable. A typical industry book is $40-$50, and a typical text book is $100-$150. The authors of these books have spent time organizing the material in a helpful way that you would otherwise need to spend. For some topics though, it's not just a matter of time savings. I'm working on a topic now that doesn't really have many useful books, so I'm having to read technical specifications produced by industry groups, which lack context and are pretty opaque. I'm missing having a book that explains these ideas in a coherent fashion. I have tried a few MOOCs for professional development. I've found they can be helpful for a superficial understanding, but they don't encourage the deep understanding I get from reading through a book.

smurpy 6y ago
Two thumbs up for Refactoring by Fowler. Exactly. Even if you already know know the 300 odd refactorings he documents... The fact that his names let you talk about them with your team or more easily treat them as abstractions is a powerful win. That book is one of the best investments you can make.
kamaal 6y ago
>>Do books sometimes say things that are obvious? There's a difference between things being obvious when a concept is exposed, and being proactively aware of a concept and having it as a practice/hobby. Unfortunately the difference between these two things is huge. Books are the cheapest and yet a high quality way of learning things, getting clarity and in general being exposed to rich quality of concepts. You could sure spend wandering in the wilderness learning things by experience over years. Or you can get them learning a book. Books aren't even that expensive, given how much they give you on the longer run.
- piva00 6y ago
  I completely agree. I've learned more through books than any other medium. They take time and, for me, it took a while to realise how I best learnt. I don't like reading technical books cover to cover, not on the first read at least. I try to look through the index for concepts that catch my interest, either because I realise I know nothing except the term (so that makes me feel a bit ignorant and curious to check it out) or it's a concept I can visualise being applied more easily. Usually from there I will read something I don't know what means and go look for that. When I've scanned through a bit to catch my interest then I start reading from the first chapter. It took me years to realise I didn't have to force myself to sit down and open a book from page 1. It's a bit obvious and stupid but I believe I'm not alone in having to learn myself how do I like to learn.
kristopolous 6y ago
> A typical industry book is $40-$50, and a typical text book is $100-$150. If this is out of some people's budgets, check eBay. Also some thrift store chains have certain locations that are book focused. You can get older edition textbooks for practically free. It's not like Claude Shannon's equations have changed. The old editions are effectively just as good. I've picked up plenty of classic books, HBR, Drucker, Christensen, Moore, etc ... $2-3 in fine condition. To be honest, half the time it looks unread. Then you can donate them "back into the stream" when you're done for others.
- randomNumber7 6y ago
  > If this is out of some people's budgets, check eBay. Just google "library genisis" ;)
  - kristopolous 6y ago
    Back in the 90s tech futurists were talking about having "the library of congress on the size of your finger nail". Now we actually could do something in that realm (LoC is about ~10tb and there's 1tb microsd cards; we're probably talking "within 5-7 years" without any exaggeration) but there's so many courts and laws stopping it from happening, the great cambrian explosion of knowledge has been held back.
    - dredmorbius 6y ago
      Every local library could store the full digitised book contents of the LoC[1] for about $5,000. That this isn't happening is a crime. ________________________________ Notes: 1. I get about 200 TB based on 40 million books, OCRd at 5 MB per book. A 2 TB SATA 5200 RPM drive runs about $100, and the full collection would fit on 50 such drives. Full optical scans are ~10x larger. The LoC's collection includes another 130 million non-book items. https://www.loc.gov/about/general-information/ https://www.loc.gov/about/general-information/
      - kristopolous 6y ago
JackFr 6y ago
I had the reaction when reading a preprint of Domain Driven Design. It was packed with ideas which were just coming into focus in my own mind after 10 years in the industry, but I didn't have names for them or have them systematized. It's a terrific feeling reading a well-written book like that and going 'Yes! Yes! Yes!'. (Since then, as with many things in our industry, it seems the book has been turned into a buzzword consulting/training money grab. Sigh.)
- whitepoplar 6y ago
  The book itself has turned into a consulting money grab, or DDD as a concept?
  - gligorot 6y ago
    Seconding this question as a reddit remindMe.
  - JackFr 6y ago
    It looks like (to me) there is a DDD training industry, along the lines of, or as a part of the “agile transformation“ training industry. Which in and of itself isn’t bad I suppose, but part of me feels that these are cynical attempts to unscrupulously monetize ideas who’s time has come by bombarding credulous middle management with buzzwords.
- jfaerge171 6y ago
  When I first read the DDD book almosty 15 years ago, I think it took me 2 or 3 reads to get it into my head. I really like the first half that can be used purely for a design perspective, and the second half for communicating with the business people to do business domain modelling . Other books I really like are The Design of Everyday Things by Don Norman, it is a good read and lets you talk to a lot of UX'ers as I have found more than a few that have used this book for their thesis. Enterprise Integration Patterns by Gregor Hohpe and Bobby Woolf, which is old but in world of streaming procesing a lot of the patterns can be reused. The Site Reliability Engineering books or their free counter parts found on https://landing.google.com/sre/books/ https://landing.google.com/sre/books/ edited to add a couple of newlines.
jedberg 6y ago
> For some topics though, it's not just a matter of time savings. I'm working on a topic now that doesn't really have many useful books, so I'm having to read technical specifications produced by industry groups, which lack context and are pretty opaque. I'm missing having a book that explains these ideas in a coherent fashion. I know a couple of tech book authors, and this is how they all got started on their first book. :) There was no good book, they did all the work of becoming an expert, and then curated their learnings into a book for others.
- intricatedetail 6y ago
  There are two groups of engineers - if they embark on opaque documented problem - one will write a book and share everything they learned, the other - patent the ideas and make sure nobody else uses them without paying.
zibbity-rw 6y ago
Yesss! Feel the same way about books on topics like Deep Learning, django, etc. Book form is much better.
tzs 6y ago
> Can you get the material for free? Almost always, yes. But it takes time to find the right material, and our time is valuable. [...] The authors of these books have spent time organizing the material in a helpful way that you would otherwise need to spend. This is one of the best things about books. Yes, I can find all the material on the web, sometimes even all at the same site. But web writers usually cannot resist the temptation into putting in a ton of links, turning the material into a directed graph. There will usually be many ways to reach a given node, some that reach it before you have read all the prerequisites. There is often no good way to know if you've actually visited all the nodes. You might have taken a fork 10 nodes ago, and the other branch goes to places that nothing in your branch links to. Unless you remember that you need to backtrack 10 nodes and try the other fork, you miss all that. I want the author to figure out what order I should visit the nodes to get everything there in a good order for learning, and then provide a way for me to effortlessly follow that order. With a book, that's easy: start at chapter 1 and read to the end. A web site could do that, but for some reason most do not.
Breza 6y ago
I completely agree with everything you say here. Well put.

smurpy 6y ago

Two thumbs up for Refactoring by Fowler. Exactly. Even if you already know know the 300 odd refactorings he documents... The fact that his names let you talk about them with your team or more easily treat them as abstractions is a powerful win. That book is one of the best investments you can make.

kamaal 6y ago

>>Do books sometimes say things that are obvious? There's a difference between things being obvious when a concept is exposed, and being proactively aware of a concept and having it as a practice/hobby. Unfortunately the difference between these two things is huge. Books are the cheapest and yet a high quality way of learning things, getting clarity and in general being exposed to rich quality of concepts. You could sure spend wandering in the wilderness learning things by experience over years. Or you can get them learning a book. Books aren't even that expensive, given how much they give you on the longer run.

piva00 6y ago
I completely agree. I've learned more through books than any other medium. They take time and, for me, it took a while to realise how I best learnt. I don't like reading technical books cover to cover, not on the first read at least. I try to look through the index for concepts that catch my interest, either because I realise I know nothing except the term (so that makes me feel a bit ignorant and curious to check it out) or it's a concept I can visualise being applied more easily. Usually from there I will read something I don't know what means and go look for that. When I've scanned through a bit to catch my interest then I start reading from the first chapter. It took me years to realise I didn't have to force myself to sit down and open a book from page 1. It's a bit obvious and stupid but I believe I'm not alone in having to learn myself how do I like to learn.

kristopolous 6y ago

> A typical industry book is $40-$50, and a typical text book is $100-$150. If this is out of some people's budgets, check eBay. Also some thrift store chains have certain locations that are book focused. You can get older edition textbooks for practically free. It's not like Claude Shannon's equations have changed. The old editions are effectively just as good. I've picked up plenty of classic books, HBR, Drucker, Christensen, Moore, etc ... $2-3 in fine condition. To be honest, half the time it looks unread. Then you can donate them "back into the stream" when you're done for others.

randomNumber7 6y ago
> If this is out of some people's budgets, check eBay. Just google "library genisis" ;)
- kristopolous 6y ago
  Back in the 90s tech futurists were talking about having "the library of congress on the size of your finger nail". Now we actually could do something in that realm (LoC is about ~10tb and there's 1tb microsd cards; we're probably talking "within 5-7 years" without any exaggeration) but there's so many courts and laws stopping it from happening, the great cambrian explosion of knowledge has been held back.
  - dredmorbius 6y ago
    Every local library could store the full digitised book contents of the LoC[1] for about $5,000. That this isn't happening is a crime. ________________________________ Notes: 1. I get about 200 TB based on 40 million books, OCRd at 5 MB per book. A 2 TB SATA 5200 RPM drive runs about $100, and the full collection would fit on 50 such drives. Full optical scans are ~10x larger. The LoC's collection includes another 130 million non-book items. https://www.loc.gov/about/general-information/ https://www.loc.gov/about/general-information/
    - kristopolous 6y ago

JackFr 6y ago

I had the reaction when reading a preprint of Domain Driven Design. It was packed with ideas which were just coming into focus in my own mind after 10 years in the industry, but I didn't have names for them or have them systematized. It's a terrific feeling reading a well-written book like that and going 'Yes! Yes! Yes!'. (Since then, as with many things in our industry, it seems the book has been turned into a buzzword consulting/training money grab. Sigh.)

whitepoplar 6y ago
The book itself has turned into a consulting money grab, or DDD as a concept?
- gligorot 6y ago
  Seconding this question as a reddit remindMe.
- JackFr 6y ago
  It looks like (to me) there is a DDD training industry, along the lines of, or as a part of the “agile transformation“ training industry. Which in and of itself isn’t bad I suppose, but part of me feels that these are cynical attempts to unscrupulously monetize ideas who’s time has come by bombarding credulous middle management with buzzwords.
jfaerge171 6y ago
When I first read the DDD book almosty 15 years ago, I think it took me 2 or 3 reads to get it into my head. I really like the first half that can be used purely for a design perspective, and the second half for communicating with the business people to do business domain modelling . Other books I really like are The Design of Everyday Things by Don Norman, it is a good read and lets you talk to a lot of UX'ers as I have found more than a few that have used this book for their thesis. Enterprise Integration Patterns by Gregor Hohpe and Bobby Woolf, which is old but in world of streaming procesing a lot of the patterns can be reused. The Site Reliability Engineering books or their free counter parts found on https://landing.google.com/sre/books/ https://landing.google.com/sre/books/ edited to add a couple of newlines.

jedberg 6y ago

> For some topics though, it's not just a matter of time savings. I'm working on a topic now that doesn't really have many useful books, so I'm having to read technical specifications produced by industry groups, which lack context and are pretty opaque. I'm missing having a book that explains these ideas in a coherent fashion. I know a couple of tech book authors, and this is how they all got started on their first book. :) There was no good book, they did all the work of becoming an expert, and then curated their learnings into a book for others.

intricatedetail 6y ago
There are two groups of engineers - if they embark on opaque documented problem - one will write a book and share everything they learned, the other - patent the ideas and make sure nobody else uses them without paying.

zibbity-rw 6y ago

Yesss! Feel the same way about books on topics like Deep Learning, django, etc. Book form is much better.

tzs 6y ago

> Can you get the material for free? Almost always, yes. But it takes time to find the right material, and our time is valuable. [...] The authors of these books have spent time organizing the material in a helpful way that you would otherwise need to spend. This is one of the best things about books. Yes, I can find all the material on the web, sometimes even all at the same site. But web writers usually cannot resist the temptation into putting in a ton of links, turning the material into a directed graph. There will usually be many ways to reach a given node, some that reach it before you have read all the prerequisites. There is often no good way to know if you've actually visited all the nodes. You might have taken a fork 10 nodes ago, and the other branch goes to places that nothing in your branch links to. Unless you remember that you need to backtrack 10 nodes and try the other fork, you miss all that. I want the author to figure out what order I should visit the nodes to get everything there in a good order for learning, and then provide a way for me to effortlessly follow that order. With a book, that's easy: start at chapter 1 and read to the end. A web site could do that, but for some reason most do not.

Breza 6y ago

I completely agree with everything you say here. Well put.