Date: Wed, 13 Mar 1996 23:40:10 -0700 From: Nate Williams <nate@sri.MT.net> To: invalid opcode <coredump@nervosa.com> Cc: Nate Williams <nate@sri.MT.net>, freebsd-chat@freebsd.org Subject: Re: To pick a (perl) fight! Message-ID: <199603140640.XAA08718@rocky.sri.MT.net> In-Reply-To: <Pine.BSF.3.91.960313214109.1476A-100000@nervosa.com> References: <199603140434.VAA08632@rocky.sri.MT.net> <Pine.BSF.3.91.960313214109.1476A-100000@nervosa.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> > I can also do a quick 'thumb' through the book looking for something a > > couple orders of magnitude faster than you can with online books. After > > That's a matter of if I can type faster than you can find the page. The > way you find the page would be to either flip through the book until you > see the familiar pages around that section, or you use the index. Right. > While > that may be all well and good, I can still lexically find it faster than > you. If you know *everything* you are looking for. If you are working off memory or looking for something similar, you will end up looking at *all* of the material that has similar contenct lexically to what I want, which in reference manuals tend to be a *LOT* of material. > For instance, I can leap to the index, click on the correct letter, > find the section I need and click on it, boom I'm there. The text is the > same, the illustrations are the same. I challenge you to a race. I *know* I can find information faster than anyone using on-line docs. Again, I've been reading for 23 years, and I read alot. This means that I have a very effecient system of finding information that is orders of magnitude faster than anything I can do on-line. You can't spead read as well with on-line books because it takes more effort to 'turn the pages' than with phycical medium. > > I've read a book, I remember what I'm looking for, but I don't remember > > where it's at. With on-line books, I spend more time looking for things > > that are related slowly than I do finding things that are related. > > That's a matter of subjectiveness and reading style. Other's may do the > exact opposite. The folks who I've talked to agree with me. Again, it's still subjective, but I've yet to find someone who prefers on-line docs to printed docs given a choice between the two. > > I have yet to find a hyper-text book that as useful to me as the > > completely random access method I use with books. And, I've found that > > Well, first off, if we are using random access methods to read the > "book", than it would probably be more accurate to call them reference > manuals (just a small semi-moot point). The first time through a book most folks skim all of the material, so you end up hitting all of the highlights. In some ways, hyper-text manuals are often more difficult to 'read' because it's hard to come up with a good balance of links. Too often you have to 'link' around the doc to find necessary information, rather than having it laid out in a more logical fashion. It's nice to be able to see refernce on how the VM system affects memory usage, but sometimes it's distracting when you don't care about memory usage, but you are trying to find out something important about the VM system. The content behind hyper-text links are always obvious, so you can never tell if it's directly relevant to the material you are reading or indirectly related, so I often spend alot of time looking for the right information in the wrong places. Again, I find the most useful function reference manuals is when I know just enough to be dangerous, but not enough to have the entire answer. When I know everything, I can look directly in the index, and and on-line manual is harder to get to because I have to load the new CD in the driver from the previous one (cause it's on a different subject, etc..). It's alot easier to grab a book than it is to load up a new CD, start up the browser software, etc... My browser software for books is always running, so it's less effort. > Now the fact is, with a robust > and efficient, yet simple search engine, I can find the information I > need, fast. I don't believe there is such a thing. Either you limit your search which requires alot of thinking on your part up front as to the 'wording' of your query, or you make your search so generic that you end up looking through alot of irrelevant information trying to find valid information. You can't have it both ways. > > Do you agree that a minority of the folks in the world have ISP access? > > No, I did not mean the world, I meant people who had internet access. You argued that anyone doing Perl development probably has internet access. I'm refuting that arguement. If people use an ISP, they inherently have Internet access, hence the term 'Internet Service Provider'. > Of the people who do have internet access, a significant majority do > have web access. There is no reason we cannot provide the html version > of the manuals on the local machine anyways. This would effectively > eliminate any argue of web access for manuals. But another useful > feature of online manuals would be the real-time updating and/or > corrections of them. For people to read manuals, they must learn 'yet another' piece of software? This isn't productive. I *still* can't figure GNU-info out, so I've basically given up trying to read the docs from them simply because the user interface sucks. Again, it boils down to having information available to the end-user that they are most comfortable in. Currently that is written docs, not HTML/SGML/Tex/Hyper-Card/Win-Helo on-line docs. > > All of the stuff I use significantly gets printed out. Printing out > > Hyper-linked pages is a *pain*, since it requires a *bunch* more work > > than it needs to be. > > Yes, I tend to agree with you there. But that is a limit of the client. > One could easliy implement an html-spider -> print client. But to do that means that the book > > > The Java tutorial is a good example of a very nicely setup Hyper-linked > > document, but once you've gone through it one the links are more of a > > hinderance than a help. > > How are they more of a hinderance? Because information is 'hidden' behind the links, and while the inter-dependencies are great and wonderful for linked docs, they are confusing and difficult to follow for general reference material. The link which talks about both the memory and VM system doesn't belong in either the VM section or the memory section but is seperate. Should it get a new chapter? There are lots of 'misc' links that get left out which contain relevant information, but it's not laid out in a format that I find useful. > > Again, this is my opinion, but in talking to others I'm finding a lot > > more people who agree with me than disagree. The paper-less society > > won't happen until you can find a way for me to have the same sort of > > speedy random-access ability that I have with a book. (I doubt you'll > > find it, but I've learned to never say never.) > > We already have faster random access than human's can provide. From > above, a robust search engine will bring you to your information faster > than humanly possible. Having more information than I could ever consume hasn't been a problem since time began. We aren't any better off now than we were 2000 years ago, and in many cases we're in worse shape because we are expected to consume and understand *MORE* information than humanly possible already. I've got plenty of information that I already need to know but don't have time to learn. Provide me information that is easy to understand that takes *less* of my time to read, not more. Time is something I'm lacking, not new (slower) ways of getting it. Nate
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199603140640.XAA08718>