where physics and life collide
Posts tagged CDS
CERN Document Server
CERN Document Server
2
1 year ago
by biglove
in You're doing it wrong
CERNLover lots-o-love expertly commented on one of CDS’s completely terrible web pages. Today I am reviewing a true classic: http://cds.cern.ch, none other than the CDS home page itself. I should admit that I feel a bit sheepish today; finding egregious flaws on CDS web pages is like shooting fish in a barrel. It is just so universally bad that I’m not sure where to start. Might as well start at the top.
click to see annotations
CDS stands for CERN Document Server. It’s a long and difficult acronym (no, not really), so the CDS folks have helped you out by putting that title right at the top where you can’t miss it. Twice.
Sandwiched between them is the tiny word “Home”. You’d be forgiven for thinking it’s a link; it’s not (I know you already clicked on it). Oh no, if it were a link, it would be a very slightly darker shade of blue indistinguishable to the human eye. Nope, “Home” is just a tiny word surrounded by ample whitespace, serving no purpose whatsoever. It’s true that that on other CDS pages, that space is occupied by breadcrumbs telling you your location, but on the main page it just looks dumb.
Let’s move on to the proud proclamation of how many records CDS has. Actually there are two such statements, about a centimeter apart on my screen … and they don’t match. Just when you thought 900,000 was an impossibly large number, well GUESS WHAT, 1,017,486 is even bigger! I guess we can never really know just how many records CDS has, but rest assured, it’s a gigantic number! A large portion of these records are “fulltext“, a term I (a native English speaker) and my English-speaking friends have never heard in our lives. I’m certainly ready to believe that it is some computer science or librarian term, but I question whether any users actually know or care what it means. Apparently it’s pretty damned impressive though for archives to be sporting fulltext, because it deserves its own sentence. As soon as I find out what the fuck fulltext is I’m going to convert all the documents on my computer to it.
Search or browse? You decide! It’s not clear from first glance what the difference is. I know what “browse” usually means on a website (basically looking through categories instead of text search), but not here. Go ahead, type in a word, click “Browse” and see if you can figure out what the fuck is going on.
Not entering any search or browse terms today? Then perhaps CDS can interest you in two incredibly dense columns filled with terms nobody understands, but by God, every single one of them is a hyperlink pointing somewhere. These columns are labeled “Narrow by collection” and “Focus on“. Every time I read these two labels, my brain grinds to an infuriated halt. Don’t those mean the same thing?! Is there a discernable reason that there are 5 checkboxes next to the left column but not the right? Oh God I am so confused. Those checkboxes and underlining under every single word are making me all misty-eyed remembering my first day learning HTML. Seriously though, the point here is to be impressed by the sheer number of subjects CDS has in its archive. You’re not supposed to be actually reading those, you idiot!
If you do attempt to read the headache-inducing arrangement of subjects, you’ll find some oddities that will make you completely lose whatever faith you may have had in CDS up to this point. For example, there are both “videos” and “videotapes”. I have to admit I don’t understand the difference here. And if you click on “General Talks”, you get a list of … videos.
A truly baffling item at the end of the second column is the heading “Archives”. Holy hell! Do you mean that up until now I haven’t been looking at archives? I thought CDS was by definition an archive! And by the way, CERN Archives apparently make up a subset of Archives, even though I thought the C in CDS stood for CERN. Actually I’m certain it does.
Finally, at the bottom of the page, far, far away from the search box, are some more search options. If you are lucky enough to know what the hell these things are (KISS, anyone?), you might want to include them in your search (after you’ve already done searches that didn’t work, I suppose). That’s kind of like Google saying way at the bottom of their page “Didn’t find what you were looking for? Would you like to search the WHOLE internet? Because up until now we’ve just been fucking with you. Step right up and click a bunch of checkboxes, and we’ll be on our way!”
Amazingly, I only have one suggestion that will fix everything at once with one stroke. Just change the word “site” at the bottom to “shite”.
I would like to apologize to the CDS developers for ridiculing their life’s work …. but, damn. You would think they would care a little bit more about their public image.
Where documents go to die
12 years ago
by lots-o-love
in Infrastructure
The CERN Document Server or CDS is a piece of software only its coder could love. For the most part it fulfills its function of storing and organizing documents, and it is clear a lot of work has gone into it. But, sadly, I can’t recall ever thinking to myself “wow, this really makes my life easier.”
In the past I have never had good luck searching for documents. Even if I know that such a document exists. Even if I in fact authored the document. Even if I personally submitted the document, I have still had trouble finding the document. One solution, at least in the last case, is to use the “Your Submissions” page associated with your account. Here is roughly what you will see if you’ve submitted a note for approval,
If you click on the image and pop-up the full sized version you will be presented with a perfect reproduction of what the most critical part of this page looks like rendered in almost any browser. Go ahead and try to read that grey line, the characters are no more than 6 pixels high. The status column in this case reads “finished.”
Anticipating that the user will likely have hundreds of submissions, all of which she demands appear on one page and yet none of which she actually wants to be able to read, the designers have decided to employ the smallest font a browser can legibly render. Then they shrink the most critical information, the line listing the actual data such as document approval status and identification numbers, just a little bit more (technically, with HTML’s <small> tag).
This text, which is not quite entirely illegible, might be forgiven if it were not for the fact that not a single bit of data on that 6 pixel high line is linked to anything. At the very least you would think that the reference number would link to the full document record, and it certainly would be handy if the action and status values linked to more details about each. Instead, after waving my mouse around in utter disbelief every time I find myself on this page, I end up squinting as I select and copy the document reference number.
The search box is only located on a dedicated search page; because, really, who needs easy access to searches on a document server? I go find the search box and paste in the reference number.
This is the point at which one starts to feel a little cocky. As you paste this nice unique identifier into a box on that dedicated search page it seems as if the whole world is finally now wrapping around your finger. And in that moment between mouse-down and mouse-up, as the shaded relief of [search] inverts and reverts, it seems this “document server” is now finally, inevitably forced to unlock it secrets, maybe to even “serve a document” to you. But, alas, if you are like me and don’t always suffer through the fame and glory of publishing public documents and instead are content to be the only person to read and reread your internally published, verbose and yet highly specific yammering, then at this point you are greeted with
No public collection matched your query. If you were looking for a non-public document, please choose the desired restricted collection first.
WTF?! To see the list of “Your Submissions” I obviously am logged in. Why the fudge can’t CDS figure this out and search every damn collection that I have access to? Instead, even though I am the holder of both an authenticated session and a unique identifier for a document which I submitted myself, I am now forced to scroll through approximately 350 collections and make a guess as to which one my document is in. Thankfully, it is not hard to identify a few good possibilities, but these collections either need to be searched one at a time or each selected one at a time in separate pull-down menus before doing the search.
Feedback