Welcome to WebmasterWorld Guest from 126.96.36.199
Forum Moderators: phranque
PDF is meant to be printed out, and read like a REAL book! ;) You can distribute it on the web, but reading it online? that's not what it was designed for. Fonts look bad and unclear, and its very cluncky.
To me Adobe has spun the functionality of PDF way beyond its unqiue function and intention, to grab some of the Web business. Ive always been perplexed why URL's (not internal anchors which make a lot of sense) are hyperlinked in PDF documents. PDF just does not have the flexibility and fast loading attributes of HTML and other formats to be used a a basis for "browsing".
Personally, my pet peeve is the fact that IE automatically opens pdf files within the browser window (when the user has Adobe Acrobat or Adobe Reader installed) rather than prompting the user to either open the file in the browser, or to save it to the hard drive first.
Only about 2% of Google's index is in pdf format and they are mainly used for:
- Lengthy second tier back-up info
- Research papers
- (Email) quotations
With three main objectives - equal printing format, secured info and small(er) file size.
In other words - nearly nobody really uses it for primary online presentations anyway.
In our own site we make clear what are pdf links so people can decide to right click and save.
Trust me, a text link with "Right click to save", along with detailed instructions on what precisely the difference is between a left and right click, still results in countless people doing a regular left click. These people then email to complain because their computer "freezes" when trying to load a lengthy ebook on their 5 year old computer through IE.
And unfortunately, there is nothing you can do to code the link so it will save the file (or even prompt to save) rather than automatically open it in the browser window. As I said, it is a major pet peeve of mine ;)
My stupid question of the day:
How can Google know what's in those things, since they're not normal markup? Is it just whatever has Monika's or Kleinberg's name on it, or is it all PR, inbound link text and page titles altogether, to the exclusion of everything else?
To a large extent i think it's the webmasters responsibility, and should be part of their usability concerns, to make sure that their visitors get their info in the easiest and most effective way.
SEs should do the same and i think it is pretty obvious when a SERP is a PDF file in google. It does not let you know of the implications. Maybe a small mouseover can inform clickers like "this (pdf) (word) (flash) document may open in it's native application and take longer to download" etc.
What we do on our site, so users get happy:
Most of our documents published (I mean, longer documents, meant for online/offline reading and/or printing) are made in two versions - HTML (better for online reading and search within), and PDF (for downloading/saving/printing). They have a little icon on the side - HTML is IE gif, and PDF is Acrobat Reader's icon gif. Also, when user "mouseovers":) a link, which leads to a PDF file, a hint like "PDF file" pops-up (or sometimes like "PDF file, 255 kB") [ a href="somefile.pdf" title="PDF file" etc. :-) ]
I think this improves a lot the usability of our site.
Experienced users can rightclick the link and save the file first, if they want; non-experienced at least will have the info about the file and its size, so if they want, they can decide no to click on the link; finally, users, which want only to browse through the document, can select the HTML version, and those wanting to print the document later, the PDF version.
Hope this little hint helps:)))
As I was reading the article it struck me you could replace PDF with www*macromedia.com and have a whole new article ;)
Right on the money - except he won't be biting the hand that's feeding him at the moment:
Maybe this is a warning shot across Adobe's bow - letting them know they really need to hire a "usability expert".
Its the best way to present an "original" scan of a document. Many sites use strange java based "tif" viewers in lieu of publishing a downloadable PDF-- they are much worse then Acrobat when it comes to usability IMO.
Or am I missing out on another way to publish the "originals"? Anyone have any suggestions?
header("Content-Length: " . filesize($filepath));
header("Content-Disposition: attachment; filename=" . $filename);
And then send the file.
I might be wrong, but i think there is a document.Save and even a SaveAs method in VBscript. I think it might be employed like this, but perhaps some experimentation is required:
or perhaps even: document.execCommand('SaveAs','pdf','doc.pdf')
Of course, only IE browsers understand this.
[edited by: claus at 1:29 pm (utc) on July 15, 2003]
For me the format is perfect, provided that you download it (instead of opening in a browser window), print it, then read it. That's what it was developed for.
But there are purposes (documents that are typically used as a printout, e.g. data sheets) where PDF leaves HTML pages in the dust where usability is concerned.
What I think all web sites using PDF documents should do:
- use the PDF symbol to alert users that a link does lead to a PDF doc
- mention the file size.
I have got a 264 kByte technical paper in PDF format on my personal site, linked from a HTML page containing much the same content. The PDF version is much better to use a printout, and it is also much smaller than the Word file it was made from (in my experience this is particularly true for graphics-heavy Word docs.)
According to logs a proportion (about a fifth= of downloads are aborted. I blame the lack of a visual progress indicator in the Arcobat client for that: narrowband users see a white window for quite a few seconds without realizing that the Acrobat client works normally.
Marcia: How can Google know what's in those things, since they're not normal markup?
Marcia, PDFs have text which is searchable, and Google can read this text (much like DOC and other non-HTML file types.
Personally, I love PDFs for very specific uses. An online brochure or product sheet is a good example - a visitor can open or download the file and print a precisely formatted copy that looks exactly like what the designer intended.
I'm making increasing use of PDFs. Instead of mass-printing color flyers (that will be obsolete before they are used up), we now design each flyer as a PDF - internally, we print out color laser copies on high-quality stock for use in quotes, etc.; we can put the same document on the web for easy access by web visitors. The best part is that it is never obsolete - as things change, it is trivial to republish the PDF.
I certainly don't see PDFs as a substitute for web pages, but they definitely have their place.
I read this article yesterday and like many of his other articles my first thought was "that makes sense" and my second thought was "that's because it's ******* obvious!"
Often, text in a pdf file is unreadable onscreen, say using Acrobat Reader because either Acrobat Distiller or Ghostscript created the file with Type3 fonts instead of Type1. And, Acrobat does a really poor job of displaying Type3 fonts *onscreen*. It's fine when printed out, but yes, it's a pain for those who want to simply preview, or even read a significant portion of the file onscreen.
This is a frequent gripe among the typesetting community (esp. TeX users).
PDFs are good for certain things...
I agree. A good example is a downloadable User Manual. This is the kind of thing that needs to be printed and kept handy.
Many lengthy reports that are meant to be read in hard copy by virtue of their size, like the BBC PDF documenting their website redesign.
But there are many many instances where a document shouldn't be pdf'd. A good example is an online version of a newsletter. It's either laziness or technical ignorance to take a one page document and convert it to pdf instead of copy-pasting it into a web page template.
I don't know if I could've done a right-click and saved or not, because I didn't try, but it irritated me that I had the document open in my browser and then couldn't save it to read later. I had to zoom in until it had scrollbars all around and then scan it.