|
Before there was a Google, it was already determined that the UI needs to react within half a second to be percieved as "responsive". People will accept a longer wait for search if search is not your core-selling point, and if there is some feedback.
In other words, it is old news 
Bastard Programmer from Hell 
If you can't read my code, try converting it here[^]
"If you just follow the bacon Eddy, wherever it leads you, then you won't have to think about politics." -- Some Bell.
|
|
|
|
|
thank you,
I knew there was a better way to explain the mess I was trying to get out of my head.
Responsiveness
Like acknowledge to someone that you are thinking about their question, "uh huh, let me think here", instead of just vaguely staring at them because you drifted off and are still catching up with what ever actual question they asked instead of the 10 minutes of unrelated gab talk.
|
|
|
|
|
There is no global measurement for load time... what you have to look for is response time...
For instance - a user profile page should response immediately and load within a fraction of second.
A financial report on the other hand will take 20-40 seconds to process, so that is the loading time... but the response should be there from the very beginning until the end (progress bar)...
The main idea is not the let the user to wait without response, that makes clear what the status of the current request...
So actually the time limits should be set per page and process... And you may come up with creative ideas around long running process (for instance we use SignalR to follow the progress of a long running process, like reports, while the user can go on and continue working on other things)...
"The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge". Stephen Hawking, 1942- 2018
|
|
|
|
|
People want searching to be fast as well, but I think people are more likely to tolerate slower searching if they get good results. I think the main issue is just clicking about the site. If the home page is slow to load, the the category slow to load, then the product slow to load people may just abandon your site and go elsewhere. Especially as there is so much competition these days, users rarely have motivation to use your site especially.
Unless every page is doing major data retrieval and analysis there is no reason for any site to take as long as two seconds for a page. You either have bad hardware or very bad coding with no thought to caching etc.
|
|
|
|
|
I don't mind pages loading slowly As Long As image boundaries are preallocated, and there's no dynamic sh1te going on that makes content that's already loaded dance.
I'm sick to death of clicking the wrong links because the links I wanted moved as I clicked them.
I wanna be a eunuchs developer! Pass me a bread knife!
|
|
|
|
|
Mark_Wallace wrote: I'm sick to death of clicking the wrong links because the links I wanted moved as I clicked them.
Honest - this is the only reason I've ever clicked on an ad since I've been on the web (circa 1993).
|
|
|
|
|
dandy72 wrote: this is the only reason I've ever clicked on an ad since I've been on the web (circa 1993).
Don't tell 'em that, for God's sake -- they'll start planning ad positioning to do just that.
I wanna be a eunuchs developer! Pass me a bread knife!
|
|
|
|
|
Now there's a horrible thought...
|
|
|
|
|
dandy72 wrote: Now there's a horrible thought...
Kinda my speciality.
I wanna be a eunuchs developer! Pass me a bread knife!
|
|
|
|
|
When you have Google's resources at your disposal, then sure, why not aim for 1 second. Until you have that budget however...yeah, people might have to wait a little bit longer.
|
|
|
|
|
I aim for "as fast as possible" with any "special performance measures."
It's cool that a second is the upper limit, but if I have a report and I can't for the life of me get it faster than five seconds then that's obviously what I'm settling for.
So how will I know if I need to use special measures to get stuff faster?
When I test my own sh*t over and over and the loading times are getting on my nerves I know I'll need to revise some code...
|
|
|
|
|
Sander Rossel wrote: if I have a report and I can't for the life of me get it faster than five seconds
I don't think that wait time is relevant, the page should respond to the user action with a status bar which should count as load time.
What pisses me off is when one of my reports never returns the data!
Never underestimate the power of human stupidity -
RAH
I'm old. I know stuff - JSOP
|
|
|
|
|
Mycroft Holmes wrote: I don't think that wait time is relevant
Depends on how often the report is used and how it's presented 
Point is, I try to get any page as fast as possible, and if that's five seconds that's five seconds.
If those five seconds piss me off because I have to load it a lot (or are obviously unacceptable) then I'm going to look for special measures such as caching, reusing "expensive" variables, minimizing roundtrips to databases and other services, threading, etc...
Of course I'm a trained professional who can see possible performance bottlenecks ahead of time, so I probably do some optimizations by default.
In a few cases I've had reports that I really couldn't get any faster.
One time I went from 15 minutes (some horrible code that wasn't mine) to three seconds, which turned to about 15 seconds after an update (again, not my code).
Getting those 15 seconds to three again required extensive database work, so we left it at 15 seconds.
15 minutes was absolutely unacceptable, especially since the whole system slowed down for those 15 minutes.
15 seconds was a lot better, but just something we couldn't improve any further without considerable time and effort.
Mycroft Holmes wrote: What pisses me off is when one of my reports never returns the data!
Why does my user report only show one user, Mr. Timeout Exception!? 
Been there, done that... 
|
|
|
|
|
Reminds me of my fist contract, back in the 90s, remediate an Access daily report that was taking 25 hours to produce
Never underestimate the power of human stupidity -
RAH
I'm old. I know stuff - JSOP
|
|
|
|
|
1 second eh?
I wish someone would tell Virgin Media that. They have lightening fast broadband (well, at least download anyway), but one of the consistently slowest web sites I have ever encountered. Navigation is not particularly slick either.
It goes without saying
|
|
|
|
|
My tolerance for how long a page takes to load is based on how important the page is to me.
(And wash out your mind with soap, what you're thinking isn't what I meant!)
So, CP, pages for paying bills online and checking my bank statement have considerable grace time. Pages for news articles don't, particularly as there is usually another link to try. My main goto for e-commerce is Amazon, and they are basically instant. Which is what they want, because the longer it takes for an e-commerce site to get to the "pay" button, the more chance of changing my mind. The "one-click" purchase was a brilliant move on their part to basically eliminate the pause, do I really need this, decision.
Latest Article - A Concise Overview of Threads
Learning to code with python is like learning to swim with those little arm floaties. It gives you undeserved confidence and will eventually drown you. - DangerBunny
Artificial intelligence is the only remedy for natural stupidity. - CDP1802
|
|
|
|
|
Make sure the static "html" piece loads lightning face, then use ajax to fill in the content that will take a bit longer due to DB access, etc.
|
|
|
|
|
2 seconds is considered human real time. Anything shorter is gravy. Anything longer and our attention starts to wander.
|
|
|
|
|
Users perceive "lag" after 100 ms.
Use asynchronous loading of "interesting" content ("buffering").
Search time tolerance is proportional to the objective.
Frequency of use is what dictates enhanced retrieval techniques (e.g. warehousing; BI ).
"(I) am amazed to see myself here rather than there ... now rather than then".
― Blaise Pascal
|
|
|
|
|
Yikes! I just received notice that Windows 10 wants to restart...to install an update...and "improve" my experience.
Wish me luck! Be afraid, be very afraid.
I have my PC backed up and my fire extinguisher at the ready. Here's hoping its not the much dreaded Windows 10 "October" update. Oh no, please no, say it isn't so!
The notice was short on information. Windows Update is also a bit stingy with further information. Contradictorily, it informs me "You're up to date". The history is similarly uninformative, with regard to this notice.
I've read advice from Microsoft not to click "Check for Updates", which apparently triggers some Beetlejuice-like phenomenon...not going to risk that.
So, with this update, we travel once again into the undiscovered country from whose bourn no traveler returns.
Folks, this is scary stuff, I'm talking human sacrifice, dogs and cats living together... mass hysteria kind-of-stuff!
Looks like I picked the wrong week to stop sniffing glue
|
|
|
|
|
Eric Lynch wrote: Here's hoping its not the much dreaded Windows 10 "October" update
You're totally safe, it's Novem
Sent from my Amstrad PC 1640
Never throw anything away, Griff
Bad command or file name. Bad, bad command! Sit! Stay! Staaaay...
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
Perhaps, you're right. I may be over-sensitive...I'm still recovering from trying to use the three seashells
|
|
|
|
|
Someone has been watching a lot of 80's movies on Netflix. 
When you are dead, you won't even know that you are dead. It's a pain only felt by others.
Same thing when you are stupid.
modified 19-Nov-21 21:01pm.
|
|
|
|
|
Sadly, no...simply having fun and remembering them decades later
|
|
|
|
|
OriginalGriff wrote: Eric Lynch wrote: Here's hoping its not the much dreaded Windows 10 "October" update
You're totally safe, it's Novem
Logically speaking, then, it's an April Fool joke.
I wanna be a eunuchs developer! Pass me a bread knife!
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
