|
Looks like Chris in jovial mood today!! 
|
|
|
|
|
|
You could remove / refactor about 50% of the "stuff" on every page and the site would be 100% better for it. It's got too many options and links in too many places. It's starting to look like a coral reef instead of the nice clean professional site it surely should be.
There's 42 buttons / hyperlinks on this message editing form alone and they're *ALL* visible and in my face at once. It's starting to look more than a little amateurish in this day and age.
The reputation stuff and graph is all fine and dandy in an Asperger's geeky circle jerk kind of way but a little love put into the basics of a good, modern, clean looking site is not only practical, it's long overdue.
Yesterday they said today was tomorrow but today they know better.
- Poul Anderson
|
|
|
|
|
We've been discussing this forever.
Part of the issue (that we have to get over) is that everyone wants something different. Some members want nothing - just content. Others want a million levers and pulleys. Us? Well, maybe we are a little aspergers geeky circle but your point is taken.
[Edit] How about we run a "redesign The Code Project" competition? We've tried it before with middling results, but I'm more than happy to give it another crack.cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
modified on Thursday, March 18, 2010 10:00 AM
|
|
|
|
|
Honestly what I forsee resolving this and many other issues is a bit of iron fisted dictatorship on your part. You can't ask the users about everything or you end up in a quagmire and nothing get's resolved and the UI starts getting all coral reefy trying to satisfy everyone.
If you think back over the years there have been many times when you guys made a change to howls of indignation, now I suspect even the most vocal don't even remember what they were against.
Surely it's good to elicit some ideas but at some point you have to say "This is what CodeProject is and will look like" and as long as the ideas are authentic and usable you simply can't go wrong (unless you get rid of the colour scheme, then it's a disaster  ).
Yesterday they said today was tomorrow but today they know better.
- Poul Anderson
|
|
|
|
|
Can you please let me know what you think of http://beta.codeproject.com/KB/asp/readfile.aspx[^] as a first step in cleaning up the pages. The article view page is the single most viewed page, hence I started there.
As to this edit page itself, any suggestions on how to cleanup / tidy up some of the options we have?
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
Very nice, it looks more modern and clean instantly.
One suggestion to take it further: put the discussion area on a tab as well. Not only would that further clean things up but it would make it easier to participate in discussion while flipping back and forth to the article itself.
Revision thing is interesting as well however if this is a mock up and you're just considering it I wouldn't bother considering the amount of back-end work involved to implement and potentially huge storage implications versus the usefulness.
Yesterday they said today was tomorrow but today they know better.
- Poul Anderson
|
|
|
|
|
Why has the search feature on this site always been 10 pounds of trouble in a 1 pound bag?
On a site like this with boatloads of great content it's easily, far and away, the most important feature, yet easily the most neglected as well.
Here's the latest WTF search moment I ran into:
I wanted to cite an article to someone. I knew the article was written by Heath Stewart but it was so far back in the mists of time that's about all the concrete info I had on it. I punched in Heath Stewart in the search and selected Articles / Quick Answers and hit GO and I get a bunch or articles that reference Heath in them but not one single article of Heath's. WTF?
It took me forever hitting and missing with my search terms to remember enough to hit upon the article in question.
If search is truly as hard to implement well on this site as it's seemed to be over the years, why not hand it off to Bing or Google or something?
You guys are messing around with reputation points and other crap and letting the ball drop on the fundamentals. Usually that's a sure sign of an open source project that's going no-where (new features without fixing fundamentals) but that doesn't seem to apply in this case so what gives?
Yesterday they said today was tomorrow but today they know better.
- Poul Anderson
|
|
|
|
|
Hi John,
Thanks for the gentle feedback
On the search page there is a Advanced Search link. Click that, it expands more options, and allows you to search by member name directly (with or without keywords).
Is this what you are after?
Using Google or Bing and searching by name would still provide a list of articles that mentioned Heath, but weren't specifically written by him.cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
Chris, as you know, I tend to agree with what John said. Search is very important, and what is offered isn't up to par. It should be easier to use, more intuitive, results should be observable in many ways (chronological is high on my list), and in the end the facility should be much more effective.
I forwarded my analysis of the situation to you about a year ago, nothing much has changed since; most of my observations would still be valid. I would like to include a link, but I fail to find it...
IMO using an external search engine isn't always good enough as the domain is well structured, something general-purpose engines aren't particularly good at, as your example points out.
I now suggest CP launches a competition about search specs. Not asking for an implementation, but aiming at a decent specification.
|
|
|
|
|
Chris Maunder wrote: Thanks for the gentle feedback
If I didn't care I wouldn't be so gentle.
Chris Maunder wrote: Is this what you are after?
No. I mean yes, in a sense, but no in a larger sense of usability etc etc. That's what the kids these days would call a usability fail.
Google resolves this by determining the domain space of the search terms first off, however simply adding the authors articles to the index created on their name might be a start.
Ideally there is one and only one search box on every page in the same place and what you enter into it is parsed into a domain specific search based on where the user currently is on the site, the type of keywords entered etc etc. Results for mixed domains can be grouped by domain i.e. "CP found these results in Articles...blah blah blah and these results in Authors..blah blah blah".
Yesterday they said today was tomorrow but today they know better.
- Poul Anderson
|
|
|
|
|
John C wrote: adding the authors articles to the index created on their name might be a start
This is already in place. However, one thing I will do do is increase the weighing on a member's name within the full text index.
John C wrote: Ideally there is one and only one search box on every page in the same place and what you enter into it is parsed into a domain specific search based on where the user currently is on the site, the type of keywords entered etc etc. Results for mixed domains can be grouped by domain i.e. "CP found these results in Articles...blah blah blah and these results in Authors..blah blah blah".
We definitely want to implement something like this. One other thing that we have (contrary to outward appearances) been looking into is replacing our current full text search with something other than the extremely limited SQL Full Text Index.
We know search isn't great, but having our own search allows us to better target our searches to our various systems (jobs, members, articles etc).
We need to do better and we will.cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
I've made a small but important change after discussing this issue with Dmitry.
Try now.
It's not perfect, not close. But it may be a little better.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
Cool, that certainly solves the original case I had yesterday.
Yesterday they said today was tomorrow but today they know better.
- Poul Anderson
|
|
|
|
|
High Achievers[^], Code Developer Ultimate 2010 & Database Cruncher 8.0 links. Please never mind if somebody mentioned already.
|
|
|
|
|
thatraja wrote: High Achievers[^], Code Developer Ultimate 2010 & Database Cruncher 8.0 links. Please never mind if somebody mentioned already.
And what's with the foo class? Which dll do I need to reference to use that, pray tell me? And could someone kindly give me a hint as to what exactly lorum ipsum means? I demand some answers here!
I wish I could see Chris's face when he sees this thread 
|
|
|
|
|
I am trying to decide whether to sob or bang my head against a wall.
Or maybe both.cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
No, wait.
What's all that, and I accidentally clicked the signup button. Is there a rollback if needed, or am I doomed? 
“Follow your bliss.” – Joseph Campbell
|
|
|
|
|
Same here with me!
I checked my CP settings, couldn't get anywhere this was set!
any rollback option? 
|
|
|
|
|
I can't see this any mroe so can't approve/comment on them any more.
|
|
|
|
|
Me neither but I figured that was just because under the new reputation system I'm back to silver status.
You should see them tho since your platinum
|
|
|
|
|
I'm a mentor but difficult to mentor people when you can't see the articles.
|
|
|
|
|
| Me too, I assumed this would sort itself out once the reputation system stabilised. txtspeak is the realm of 9 year old children, not developers. Christian Graus
|
|
|
|
|
Chris ran a update on the reputation system yesterday but I still can't see them now.
Maybe the conditions have changed.
|
|
|
|
|
Apparently, you have to be a Gold Author or higher to see unpublished articles
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
