|
Today CP servers are fast in responding to GET request but when messages are posted, they are terribly slow. Looks like there are bottlenecks with backend database?
Any similar experiences?
Vasudevan Deepak Kumar
Personal Homepage
Tech Gossips
A pessimist sees only the dark side of the clouds, and mopes; a philosopher sees both sides, and shrugs; an optimist doesn't see the clouds at all - he's walking on them. --Leonard Louis Levinson
|
|
|
|
|
message posting is always bottlenecked on my end.
--
Help Stamp Out and Abolish Redundancy
The preceding is courtesy of the Department of Unnecessarily Redundant Repetition Department.
|
|
|
|
|
Yes, we know.
You have access to the beta site so should be aware of the changes we have implemented there. I guess I could go tighten the rubber bands on the servers a little, but apart from the rewrite we're doing I'm not sure what else we can do
cheers,
Chris Maunder
CodeProject.com : C++ MVP
|
|
|
|
|
Hi All/Chris,
I've been tossing this issue around for a while, and I wanted to offer up some comments (I know...). I believe it will improve the quality of submissions to the CodeProject.
* Rendering
At times, the editor simply does not render the same page which will eventually be served up by the CodeProject server. Reconcilling this issue will provide the visual feedback for submission formatting errors early in the process, where they are easier to fix for the author.
* Image Base consolidation
At times, after inserting an image, the blank image (Red 'X") is displayed. I believe it is a side effect (or part of) the basic rendering problem described above.
* Force a Preview
After clicking 'Submit', offer the Preview as the next screen with a chance to go back and edit. Too many (including myself) do not take advantage of the feature. Coersing its use will at least bring to light some of the formatting errors.
* Spell Checking
Add a spell checker module. Skip the grammar (it's too hard a problem). No explainations necessary. (Lord knows I need it).
* On the wish list
Cleanup the validation code. When validating during submit, my machine basically locks because Explorer is consuming all available cycles. For me, this can last between 15 and 30 seconds. Yuk!
Jeff
|
|
|
|
|
The rendering engine is IE. If IE's WYSIWYG editor and IE's HTML viewing engine differ then that makes life difficult. Have you tried Firefox? Could you please send me specific snippets of HTML that render differently in the editor and the final article? I will investigate. I'm starting to feel we should simply move to another editor. It won't fix all problems but at least we could use something more widely used.
Image preview: Is this in the "Add image" popup? That one's a funny one. Doesn't work sometimes, works every single time I try and debug it.
Force Preview: Not a bad idea at all.
Spell check: Thought about this one too and have investigated some options but I've not found anything that makes me feel confident in it. There's always FireFox as an interim solution.
Validation: Not sure what you mean. Any delay between the final "submit" and the article being finally sumitted is server-side, not client side. Yet another of the things we're working on
cheers,
Chris Maunder
CodeProject.com : C++ MVP
|
|
|
|
|
Hi Chris,
Chris Maunder wrote: Could you please send me specific snippets of HTML that render differently in the editor and the final article?
Will do when I come across it. If I recall, it had to do with copy and paste from a third party site. The pasted code picked up attributes that were in a [SPAN] tag. In the WYSIWYG editor, the background color attribute was not rendered. When the article was received from the server, the tags were still present, so the user agnet rendered them.
I suspect this is why a recent article had a bright yellow background on the text (and not the entire body of the document).
Chris Maunder wrote: Doesn't work sometimes, works every single time I try and debug it.
Of course.
Jeff
|
|
|
|
|
Hi
I haven't been seeing this link for a couple of days now... am I missing some setting?
Thanks
Chandra
|
|
|
|
|
ChandraRam wrote: I haven't been seeing this link for a couple of days now... am I missing some setting?
Maybe it's been disabled for Bronze level members? [just a guess]
|
|
|
|
|
Nishant Sivakumar wrote: Maybe it's been disabled for Bronze level members? [just a guess]
If that were the case, does it imply that bronze members cannot think for themselves?? or were somehow inferior that their opinions need not count?
But seriously, what other features are disabled or are unavailable to bronze members?
Thanks
Chandra
|
|
|
|
|
Posting status:<br />
<br />
Lurker No Messages have been posted <br />
Poster The member has posted up to 100 messages <br />
Regular The member has posted from 100 up to 1000 messages <br />
Personality The member has posted 1000 up to 5000 messages <br />
Fixture The member has posted over 5000 messages <br />
<br />
Author status:<br />
<br />
Browser No Articles have been posted <br />
Contributor The member has submitted up to 4 articles <br />
Writer The member has submitted 5-14 articles <br />
Prolific The member has submitted 15-24 articles <br />
Legend The member has submitted over 25 articles <br />
<br />
Membership status calculated as follows:<br />
<br />
1-4 articles posted = Bronze. 5-14 articles = Silver. 15-24 articles = Gold. 25+ articles = Platinum. Each 500 messages posted adds one level, and if a user has posted a message then each year of membership adds 1 level. Bonus levels not valid for Gold members. Bronze is awarded at the beginning of the first 500 messages posted instead of the end.
I would say these restrictions should apply from the first category ('Posting Status') instead of from articles. There are a few experts who love to participate in the forums but feel a bit bored and impatient to type out a at-length article.
Vasudevan Deepak Kumar
Personal Homepage
Tech Gossips
A pessimist sees only the dark side of the clouds, and mopes; a philosopher sees both sides, and shrugs; an optimist doesn't see the clouds at all - he's walking on them. --Leonard Louis Levinson
|
|
|
|
|
It's to stop someone creating a new account, posting 1 message, then going and reporting (and hence banning) other members' posts
cheers,
Chris Maunder
CodeProject.com : C++ MVP
|
|
|
|
|
Would it be possible to make it that if someone votes for an article, it should be mandatory for them to leave a comment as to why they voted the way they did? Let's take for example this article here: http://www.codeproject.com/useritems/GenericSingleton.asp[^]
At the time this message was posted, it was rated far below than it should it have been ( <= 2.0), and yet there were no comments as to why someone would have voted that low for it. It's fairly well written, provides a novel approach to Singleton Patterns, etc. and yet whoever nuked it gave no clue as to why.
|
|
|
|
|
JohnnyLocust wrote: At the time this message was posted, it was rated far below than it should it have been ( <= 2.0), and yet there were no comments as to why someone would have voted that low for it. It's fairly well written, provides a novel approach to Singleton Patterns, etc. and yet whoever nuked it gave no clue as to why.
I am not so sure it's a novel approach. There are far too many articles on using the singleton pattern here on CP and on the web. So such articles often run a risk of getting severely low votes.
If you are writing on a very thoroughly exhausted topic, it's best to repeatedly explain at the very beginning how your article differs from previous ones.
|
|
|
|
|
So such articles often run a risk of getting severely low votes.
Did you even read what I wrote? If someone voted his article a 1 because "There are far too many articles on using the singleton pattern here on CP and on the web." how would the author know what he did right or wrong if no comments were made along with the vote?
|
|
|
|
|
JohnnyLocust wrote: Did you even read what I wrote? If someone voted his article a 1 because "There are far too many articles on using the singleton pattern here on CP and on the web." how would the author know what he did right or wrong if no comments were made along with the vote?
I agree that a comment would be helpful. But insisting on a comment for every vote is not a good idea at all. People would just be pissed. Some people will vote, some comment, others do both. We have to live with that.
In this particular case, he'd just have had to do a bit of research around CP to figure out why his article was not well received. It's one of the most elementary checks to do when submitting an article to a website to see if others have already done so.
|
|
|
|
|
JohnnyLocust wrote: There are far too many articles on using the singleton pattern here on CP and on the web
Not all fingers are same. If the author has truly intended his article from his heart and has written it (without attempting plagarism) then he should be saved from this 'Low-Vote-Menace'.
Vasudevan Deepak Kumar
Personal Homepage
Tech Gossips
A pessimist sees only the dark side of the clouds, and mopes; a philosopher sees both sides, and shrugs; an optimist doesn't see the clouds at all - he's walking on them. --Leonard Louis Levinson
|
|
|
|
|
I think thats a great idea.
Somebody votes your article down, you want to know why and how you can improve it. Its actually a little unfair to give a low vote for no apparent reason, unless they just don't like you.
My current favourite word is: PIE!
I have changed my name to my regular internet alias. But don't let the 'Genius' part fool you, you don't know what 'SK' stands for.
-The Undefeated
|
|
|
|
|
JohnnyLocust wrote: it should be mandatory for them to leave a comment as to why they voted the way they did?
If you want to fly around the world with a baseball bat and menace voters until they comment, be my guest! 
Short of that, there's really no way to make it mandatory. I mean, we could say it's mandatory, maybe throw in some checks that a non-empty comment of some sort was left, but really what's to stop someone from just typing "slkfjsdlfk" or "i don't like it." and moving on...?
---- ...the wind blows over it and it is gone, and its place remembers it no more...
|
|
|
|
|
This is an issue which no one seems to want to address. I've got a couple of fans who constanly offer univotes. I've asked for histograms to try and see the statistics, but they are not available. That leads me to believe providing a list of IPs (to investigate further) is out of the question.
It seems the site will only take action when a threshold is met in the forums, after enough readers complain. The authors of the articles to left to hang out and dry.
I'm at the point I ask if there is any interest, because I am tired of my rating getting beat up on less than main stream articles by the univoters. See ASN.1 Parsing (at Article Requests)[^]
Jeff
|
|
|
|
|
Jeffrey Walton wrote: I've asked for histograms to try and see the statistics, but they are not available.
I think in the .NET rewrite, we'll have a more detailed vote view (how many 5s, 4s, 3s etc.). That would be a better reflection of how people have rated an article.
Also, eventually anything above 4 should be considered a decent vote. Some people will not like your article anyway, so you'll always get a few 1s. I know most authors will find it hard to accept that someone would dislike their article. But that's how it is.
I am always a little amused every time I see an angry author complaining of getting 1-votes. I did that several years ago, and I have seen dozens of other authors do the same. Eventually most regular CP authors settle down to ignoring the low votes they get.
|
|
|
|
|
I'm so mad 
They didn't even leave a comment... wait, they left a comment but it was useless. Arrgg. I'll have to go puch something now, to get rid of all this pent up rage!
My current favourite word is: PIE!
I have changed my name to my regular internet alias. But don't let the 'Genius' part fool you, you don't know what 'SK' stands for.
-The Undefeated
|
|
|
|
|
Hi Nish,
Nishant Sivakumar wrote: I know most authors will find it hard to accept that someone would dislike their article. But that's how it is.
I agree. I would find it more palatable if the submission lacked quality. I've been keeping statistics with respect to what I recieve. What I've found is damn near as soon as one of mine hit, one or two univotes come in. Watching other articles submitted at nearly the same time (next to the article in question on the 10 most recent list), show others are not receiving them. This tells me it is arbirary and capricious.
Jeff
|
|
|
|
|
Jeffrey Walton wrote: This tells me it is arbirary and capricious
I should qualify this by stating the univoters are arbirary and capricious, not CP. I believe it was Marc who coined the term for this type of voter as 'fans'.
Jeff
|
|
|
|
|
Jeffrey Walton wrote: the site will only take action when a threshold is met in the forums
Jeffrey,
"The site" is me, insofar as decisions on what our team works on, and our number 1 priority is getting the new base infrastructure out the door so that we can continue to handle the growing load and not completely fall over.
Histograms will be available, we will also look to install voter tracking to weed out univoters, but they aren't going to happen today, and I am a little offended that you feel we leave authors to hang out and dry.
Please let me assure you that myself and my team are working literally around the clock and our weekends to enable us to introduce the features our readers and contributors have asked for and we are not sitting here ignoring you.
We're still a small team, though, and the infrastructure is massive, so getting the features that were requested by you and the thousand-or-so members will take time. For each feature we add we are going to have dozens of members complain that we should be concentrating on feature X instead of Y
cheers,
Chris Maunder
CodeProject.com : C++ MVP
|
|
|
|
|
Chris Maunder wrote: I am a little offended that you feel we leave authors to hang out and dry.
My apologies. My personality can be abrasive at times, with no mal-intentions.
I suppose it depends on how you look at it, but I only become passionate about things I care about. So if I did not care, my abrasiveness would not show (I know it is kind of perverted).
Jeff
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
