Is that a new design?



  • Well, I just discovered the pitfalls of browser cross-testing today, but not for the reason you would expect. I have a series of web portals utilising the same interface that we were tasked with updating somewhat as the whole organisation has just had their outdated monitors replaced - we're all widescreen now baby! Nothing particularly fancy, but it doesn't work particularly well with IE7 or IE8 due to CSS constraints - we were given a preferred template to work off, so we made it work.. A non-issue when our entire organisation is IE9+ (I checked with IT). So, off I go and test IE9, 10, 11, Chome, FF and Safari. All good! Lets go.

    Unless IT run out a group policy forcing a small AD computer container to use IE7 emulation mode. An AD container that happens to contain a lot of machines belonging to higher-ups. Queue phone calls when I deploy a change to add in a few bells and whistles at the request of a few of those higher ups - a bit of CSS jiggery pokery, nothing major but a game-changer for IE7 users. A quick investigation with the Head of IT who spotted the issue first and we push the X-UA-Compatible meta tag out for IE users and it's all resolved, but still. IE7 forced? Ouch. Of course, my machine and that of all our UAT group don't happen to be within that AD container.

    I thought I was TRWTF but it turns out I was only a minor wtf in the grand scheme of things. After all, who in their right mind enforces IE7 compatibility mode via GPO?


  • :belt_onion:

    @thegoryone said:

    After all, who in their right mind enforces IE7 compatibility mode via GPO?



  • @thegoryone said:

    After all, who in their right mind enforces IE7 compatibility mode via GPO?

    Probably someone with a mission-critical legacy web application they can't port. Same reason a lot of smelly shit is kept, really.



  • I bet that shitty "how to demoralize employees" company I worked for did.



  • @Medinoc said:

    Probably someone with a mission-critical legacy web application they can't port. Same reason a lot of smelly shit is kept, really.

    Thing is, you can set compatibility mode on a per-domain basis...so said legacy web app still doesn't excuse them turning it on globally...


  • :belt_onion:

    @tarunik said:

    Thing is, you can set compatibility mode on a per-domain basis...so said legacy web app still doesn't excuse them turning it on globally...

    Given the OP it is possible that all the portals are on the same domain.


    Filed under: I doubt it, though



  • @Onyx said:

    Given the OP it is possible that all the portals are on the same domain.


    Filed under: I doubt it, though

    They are on the same domain - single domain across 6 sites with satellite domain controllers reporting to a primary on another site. The AD containers go something along the line of Domain -> OU - (physical) site -> container type -> PC container (Technically forest but with one domain, it's basically the same thing). Most of the powers that be are based in the same site so their machines reside in the same AD container, while I work at the next largest site so mine does not.
    [quote=Medinoc]Probably someone with a mission-critical legacy web application they can't port. Same reason a lot of smelly shit is kept, really.[/quote]

    That's my assumption too, but I have access to the majority of legacy systems as I've replaced half of them by now and none of them needed any IE7 compatibility forced, so I'm frankly terrified to find out which one of the remaining ones needs it - the prime candidate being their financial web application. I'm not going anywhere near that one.

    Edit: Also, remember this is only forced on a small sub-set of users in a single PC container under a specific site in AD - so if they log into a different machine in a different location, theory suggests no IE7 forced and issues, but I've heard of none.


  • Discourse touched me in a no-no place

    @thegoryone said:

    After all, who in their right mind enforces IE7 compatibility mode via GPO?

    That's a funny coincidence, this morning we ran into the exact same problem where I work.

    I suspect it was a change made by IT because we're using a vintage version of Microsoft CRM. It's not necessary to force IE into compatibility mode though, because CRM dishes up the correct headers to tell IE which mode is required.

    We've now stuck IE=Edge in the header of the site that was giving us trouble (which I thought was already standard practice here) and it works fine.



  • @DoctorJones said:

    That's a funny coincidence, this morning we ran into the exact same problem where I work.

    //snip

    We've now stuck IE=Edge in the header of the site that was giving us trouble (which I thought was already standard practice here) and it works fine.

    Same solution, though we added IE=Edge;IE=9;IE=8 just in case. The minor WTF was an oversight in my case to overlook that something silly like this might have been needed.



  • @thegoryone said:

    I just discovered the pitfalls of browser cross-testing today

    Bah! Peanuts compared to the amount of crap I have to go through to get an app working on the different SmartTVs. Look, at least you didn't have to order some cheap ass Chinese UART so you can deploy to crappy SmartTVs using USB.



  • @DoctorJones said:

    We've now stuck IE=Edge in the header of the site that was giving us trouble (which I thought was already standard practice here) and it works fine.

    For our site, our customers started having troubles on some pages due to Compatibility Mode being globally enabled on their machine soon after IE8 came out. We did test against IE7, but some JavaScript code (events?) would act different between real IE7 and IE8/9 with compatibility turned on. Other than through GPO, I don't even understand how Compatibility Mode is enabled globally on so many home computers too.

    One day the support people got tried of answering those kinds of emails (and trying to explain how to turn it off... especially when it seemed like it would turn on for only one page of our site, but the broken page icon wasn't appearing in the address bar to toggle it, and nothing listed in the site list). So, we pushed out a change to have Apache send an IE=Edge HTTP header for everything served on our site. Problem solved!

    From my perspective, the auto-detection of when to enable Compatibility Mode was a pretty retarded move by Microsoft, along with the global setting even being possibly set. Maybe it has helped some IT department out there, but certainly not us with a public web site.


  • Discourse touched me in a no-no place

    @tarunik said:

    Thing is, you can set compatibility mode on a per-domain basis...so said legacy web app still doesn't excuse them turning it on globally...

    That depends. Is there a common timesheet web application? For years, we were stuck with that sort of thing. (Admittedly only on the departmental level, but that's a small mercy.) Well, unless you had a copy of J++ and could run the local client.<There were vendor updates but we were too cheap to buy them…



  • @dkf said:

    That depends. Is there a common timesheet web application? For years, we were stuck with that sort of thing. (Admittedly only on the departmental level, but that's a small mercy.) Well, unless you had a copy of J++ and could run the local client.

    When I said "domain", I meant "domain name of Web site", not "domain of AD clients"...



  • @dkf said:

    That depends. Is there a common timesheet web application? For years, we were stuck with that sort of thing. (Admittedly only on the departmental level, but that's a small mercy.) Well, unless you had a copy of J++ and could run the local client.<There were vendor updates but we were too cheap to buy them…

    Actually, there is. That said, It's got a meta tag set in its code and it's forcing itself to IE8 locally, so not sure if that's the probable culprit.



  • @tarunik said:

    When I said "domain", I meant "domain name of Web site", not "domain of AD clients"...

    It sure was a good decision to re-use that name<and I'm sure some pendant will show up telling me I'm wrong about which term is reused>.



  • About as good as the idea a lot of companies have to match their internal AD domain with their web domain, eh? Case in point...



  • @thegoryone said:

    About as good as the idea a lot of companies have to match their internal AD domain with their web domain, eh? Case in point...

    :wtf:



  • @thegoryone said:

    About as good as the idea a lot of companies have to match their internal AD domain with their web domain, eh? Case in point...

    I know we have multiple domains. My machine isn't Windows, so it's not on there, but I have to know it for email / LDAP purposes.

    None of our ADs' names match our web domain.

    I'm not any sort of a sysadmin, so your post is a little too cryptic for me, though given how windy it's been here lately (global warming, where are you‽), I probably wouldn't notice.



  • @boomzilla said:

    I'm not any sort of a sysadmin, so your post is a little too cryptic for me, though given how windy it's been here lately (global warming, where are you‽), I probably wouldn't notice.

    Essentially from what I recall from my days as an engineer (Though I was primarily VMware/Shitrix) it's generally considered a bad idea to use <domain>.local or your external domain name for your AD domain. It's caught on a little more these days but while it doesn't basically break stuff now like it used to, it causes extra work for sysadmins to get everything set up (split DNS issues iirc). I'm pretty sure it's generally recommended you either go the sub.domain.extension pattern (eg internal.microsoft.com) or a different domain (eg you own company.co.uk and company.uk.com, use one for your web presence and park the other and use it as an internal domain).

    We do have a second internal domain with a different suffix but everything just runs off our primary instead for some reason.



  • @thegoryone said:

    Same solution, though we added IE=Edge;IE=9;IE=8 just in case. The minor WTF was an oversight in my case to overlook that something silly like this might have been needed.

    Are you setting it in the HTML head with meta tags or the HTTP headers from the server? I highly recommend the latter if you can make changes to web.config. You don't have to worry about the position of the directive in the document that way - if you place the meta tag too low, you can cause the IE rendering engine to abort and restart, which totally kills the page performance.



  • Since the meta tag takes precedence to the HTTP headers (Found that out as part of my research) we're sitting on it for now. It's also the only system we're running with atm pushing this level of CSS3/Ajax/jQuery integration and we're seeing no issues with other existing systems. Something we'll probably take a look at if we have to push it out further (Which is possible).

    And right after the head tag. What do you take me for, some kind of walking WTF? Cos I probably am.



  • @thegoryone said:

    Since the meta tag takes precedence to the HTTP headers (Found that out as part of my research) we're sitting on it for now. It's also the only system we're running with atm pushing this level of CSS3/Ajax/jQuery integration and we're seeing no issues with other existing systems. Something we'll probably take a look at if we have to push it out further (Which is possible).

    And right after the head tag. What do you take me for, some kind of walking WTF? Cos I probably am.

    True, meta takes precedence; it's just potentially a speed penalty if it causes the render engine to abort and restart.That's all I was trying to say. It may not trigger an abort, or it may not matter for you even if it does.



  • @Medinoc said:

    thegoryone:
    After all, who in their right mind enforces IE7 compatibility mode via GPO?

    Probably someone with a mission-critical legacy web application they can't port. Same reason a lot of smelly shit is kept, really.


    Or, because there once was a couple of mission-critical legacy web applications that couldn't be ported and belonging to different departments or whatnot, and though since then both of them have been replaced / ported / whatever, the departments in question forgot to communicate this and thus everyone keeps it forced because they recall someone else needed it.

    Or maybe it's because of sheer costs of updating (packaging, engineers'
    traveling costs, work hours, bandwidth (oh, wait, they have CDs for this), licences, and then the same for the net-nanny of choice and all the plugins that absolutely need to be there because they make the bosses feel safe, as well as training costs for the employees and the expected peak in the support calls the following week.. I mean month).



  • @rad131304 said:

    True, meta takes precedence; it's just potentially a speed penalty if it causes the render engine to abort and restart.That's all I was trying to say. It may not trigger an abort, or it may not matter for you even if it does.

    Now where's the fun in life if you can't intentionally leave in a little bit of WTF for the next guy?



  • @thegoryone said:

    I'm pretty sure it's generally recommended you either go the sub.domain.extension pattern (eg internal.microsoft.com) or a different domain (eg you own company.co.uk and company.uk.com, use one for your web presence and park the other and use it as an internal domain).

    Back in my previous place we just used example.office to go with our example.com. (Not literally example I hope you understand). This was back in the SBS 2003 days.

    @thegoryone said:

    bad idea to use .local

    AFAIK certain zeroconf systems used this. If you were pure Microsoft it probably wouldn't matter but any Linux or Mac machines would cause all kinds of issues! I've always worked in heterogeneous environments so this can be a concern.


  • :belt_onion:

    @Zemm said:

    If you were pure Microsoft it probably wouldn't matter but any Linux or Mac machines would cause all kinds of issues!

    Encountered this. Linux machines complained to no end, but worked. Now, I'm pretty sure there is a usecase where everything would break horribly, but this is a company we had to fight tooth and nail to even install a Linux VM, let alone put a file share on it, regardless of test results we presented that proved a Linux SMB share will solve a problem they were facing. Making them to change the domain name is damn impossible.


    Filed under: Cue the "Everyone forgot it even exists now while they keep fiddling with Windows machines daily" cliché.



  • @Zemm said:

    Back in my previous place we just used example.office to go with our example.com. (Not literally example I hope you understand). This was back in the SBS 2003 days.

    @thegoryone said:

    bad idea to use .local

    AFAIK certain zeroconf systems used this. If you were pure Microsoft it probably wouldn't matter but any Linux or Mac machines would cause all kinds of issues! I've always worked in heterogeneous environments so this can be a concern.

    Seen it done (.local) in a lot of small organisations in the past running SBS2003 (You see how out of date my AD knowledge is? Even though I have the MCITP:EA on 2008...) but a quick glance online says MS are now recommending you don't do this now. 99% sure they used to recommend it though.



  • @boomzilla said:

    given how windy it's been here lately (global warming, where are you‽), I probably wouldn't notice



  • @Zemm said:

    AFAIK certain zeroconf systems used this

    Using .local as a fake TLD was MS-recommended best practice for a long time. Pity Apple didn't get the memo.

    In recent years, both Mozilla and Chrome browsers try to look up .local names via MDNS instead of DNS and just hang instead of falling back when they get no response. The admin.local and curric.local domains at the school now have admin.lan and curric.lan synonyms for anything webby that needs hosting locally.



  • Our domain now is notthecompanyname.private but previous domain for one of the companies we're made of matched their external .com domain.



  • @boomzilla said:

    how windy it's been here lately (global warming, where are you‽)

    Wind is caused by the warming of the Globe.

    @flabdablet said:

    In recent years, both Mozilla and Chrome browsers try to look up .local names via MDNS

    RFC 6762 I guess

    @flabdablet said:

    The admin.local and curric.local domains at the school now have admin.lan and curric.lan synonyms for anything webby that needs hosting locally.

    Pretend I work for "Wxyz company": We use whatevername.wxyz for internal names. My dev site is called "zemm.wxyz" (substitute real names), which can be a problem when testing web call-backs etc. (Luckily I have a "dev site" in the "production environment" dev-zemm.wxyz.com.au hooked up to SVN. And zemm.wxyz.com.au for an older release version)

    No AD domain due to a general anti-MS feeling and still being fairly small. But we'll probably need one one day.



  • @flabdablet said:

    <img src="http://assets.amuniversal.com/95ca932082a60132bfba005056a9545d">

    Well, that's less wrong than what you usually post, which just means you didn't draw it, I guess.

    Fortunately, last night's batch of global warming was nice and powdery and light, so it was easy to shovel.



  • @Zemm said:

    Wind is caused by the warming of the Globe.

    Yeah, but as the poles catch up (polar amplification!) there should be less difference in temperature between the poles and the tropics, so weather should calm down a bit. Plus...fucking cold wind...brrr....



  • @boomzilla said:

    Plus...fucking cold wind...brrr....

    Well it was windy here today too. Fucking hot wind!

    It is now 10:30pm and my thermometer says it is 27°C. Even after the small rain storm that just passed over.


  • Discourse touched me in a no-no place

    @boomzilla said:

    Yeah, but as the poles catch up (polar amplification!) there should be less difference in temperature between the poles and the tropics, so weather should calm down a bit.

    Maybe or maybe not. Fluid dynamics is close to the very definition of non-trivial…



  • @dkf said:

    Maybe or maybe not. Fluid dynamics is close to the very definition of non-trivial…

    True. Still, the difference in temperature is what drives weather. The "bad weather I win, good weather you lose" schtick is getting pretty old.



  • It's (or feels like) 10 degrees celcius here most of the year round and won't stop raining. We lose.


  • Discourse touched me in a no-no place

    @boomzilla said:

    The "bad weather I win, good weather you lose" schtick is getting pretty old.

    That's been a recognised straw-man for ages. The really big problem is that it changes stuff, and in difficult-to-predict ways. Will you be a winner or a loser overall? Dunno. Do want to effectively gamble everything on that? (I don't…)

    Scenarios that really scare people include key things like a drying of the world's main breadbaskets. If that happens, we're up Shit Creek, even if there's somewhere else we can transfer the production to (because that'll take a long time) and lack of food is one of the things that can trigger overall social collapse. (We know that last point from history.) Reductions in water supplies to major urban areas would be pretty bad too, and I've seen some arguments that we've started to see wars (at least partially) over water, even if I'm not sure I agree with those arguments.

    Everything most certainly can be mitigated with enough time and investment. The human race isn't doomed. Things might get pretty bad in the meantime though…



  • @dkf said:

    That's been a recognised straw-man for ages. The really big problem is that it changes stuff, and in difficult-to-predict ways. Will you be a winner or a loser overall? Dunno. Do want to effectively gamble everything on that? (I don't…)

    Oh, the precautionary principle. How can you tell what's changing what? Sorry, but this is terribly unpersuasive once critical thinking enters the scene.

    @dkf said:

    Scenarios that really scare people include key things like a drying of the world's main breadbaskets.

    Yeah, but so what? We have no skill in predicting rainfall changes WRT climate. But then, I'm not aware of anything where we have skill in predicting climate.

    @dkf said:

    Things might get pretty bad in the meantime though…

    Fuck off and come back when you have something better than a scary story, is my response.

    Climate science is pretty much trying to get into the epicycle stage of development, from all indications. At least epicycles had some predictive skill.



  • @quijibo said:

    We did test against IE7, but some JavaScript code (events?) would act different between real IE7 and IE8/9 with compatibility turned on.

    As I discovered recently when debugging problems IE11 has with my organization's main data collecting webapp, IE has two different emulation modes. For IE7, Those are IE=7 and IE=EmulateIE7. IE8's modes are similar.

    Supposedly the difference is IE=7/IE=8 not allowing Quirks mode rendering, but the one spot I ran into problems with it and had to use EmulateIE8 was because IE wasn't making an object tag with a 100% width and height fill its available space.



  • We recently had our customer turn on that group policy way of forcing compatibility mode for our app. No clue why. Certainly, no one asked them to do it. About a year ago, we needed IE9 to run in compat mode, but we upgraded some stuff and now it breaks. Also, they're finally rolling out IE11.


  • Discourse touched me in a no-no place

    @boomzilla said:

    How can you tell what's changing what? Sorry, but this is terribly unpersuasive once critical thinking enters the scene.

    The problem is if you wait until you've got concrete evidence for everything, you run the real risk of having waited until after the point where you can mitigate the problem.

    The one thing we're pretty sure about at this stage is that there's a fairly long lag time between the energy and chemical inputs to the climate system and the stable average yearly thermal equilibrium that those inputs predict, and that this is largely due to the time it takes to transfer heat in and out of the oceans. The time lag is what makes this important, yet difficult to spell out in terms of immediate consequences.

    Most of the people advocating action now do so because they think it is the cheapest, least disruptive approach with fewest possible down-sides.

    @boomzilla said:

    Yeah, but so what? We have no skill in predicting rainfall changes WRT climate. But then, I'm not aware of anything where we have skill in predicting climate.

    I have no control over what you're aware of. :smiley: Cheap shot aside, climate predictions are statistical anyway (all things to do with climate are; the detail is called weather) and there's always the worry that there's something important hiding in there that nobody's accounted for in the models. An example of this was when they discovered more about the coupling between the stratosphere and the troposphere: it turns out to be critical and to depend on features that were not modelled until comparatively recently. There's also the atmosphere/ground coupling — poorly understood at global level because of things like chemical interactions — and atmosphere/ocean coupling.

    I know enough to know I know very little. ;)

    @boomzilla said:

    Climate science is pretty much trying to get into the epicycle stage of development, from all indications.

    There are people out there trying to stir up the perception of controversy where there isn't any (probably for commercial advantage; that's the usual reason for the sorts of tactics that appear to be being deployed) and it looks like the epicentre of their activity is the US. Be very careful of what you read. When the only experts on one side of an argument have a strong pecuniary interest in one point of view and all the other experts are on the other side, there's a distinct whiff of bias out there, not indication of fundamental disagreement.

    BTW, if you've got any good ideas for technical fixes for systematically reducing CO2 levels in the atmosphere, it's a really good idea to see if they'll work and scale up. The green nutjobs might want to send everyone back to the stone age as Plan A, but everyone else thinks that they're full of shit and other approaches should be used.


  • Discourse touched me in a no-no place

    @dkf said:

    Cheap shot aside, climate predictions are statistical anyway (all things to do with climate are; the detail is called weather) and there's always the worry that there's something important hiding in there that nobody's accounted for in the models. An example of this was when they discovered more about the coupling between the stratosphere and the troposphere: it turns out to be critical and to depend on features that were not modelled until comparatively recently. There's also the atmosphere/ground coupling — poorly understood at global level because of things like chemical interactions — and atmosphere/ocean coupling.

    I know enough to know I know very little.

    All of this implies that we really don't know enough about how climate works to accurately model it, but we're not letting that stop us.


  • Discourse touched me in a no-no place

    @antiquarian said:

    All of this implies that we really don't know enough about how climate works to accurately model it, but we're not letting that stop us.

    We can wait until we're certain, and then be certainly very sorry when we find we can't fix anything because we waited far too long. :smile: Certainty is for fools.


  • Discourse touched me in a no-no place

    @dkf said:

    We can wait until we're certain, and then be certainly very sorry when we find we can't fix anything because we waited far too long. :smile: Certainty is for fools.

    If certainty is for fools, who is unjustified certainty for?

    If we don't know enough about climate to accurately model it we certainly don't know enough to change it in the way we want, or even if we can change it in the way we want. Not to mention the fact that we could be due for another ice age, which would be far worse than anything the climate change people are saying will happen. What's worse, we may not be able to do anything about it.



  • @dkf said:

    Cheap shot aside, climate predictions are statistical anyway

    Yes, I'm familiar with that dodge. But the current predictions are failing pretty much no matter how you try to rationalize them. Way too many unknowns in how the system works. And even if we knew them, we might still not be able to make good predictions.

    @dkf said:

    When the only experts on one side of an argument have a strong pecuniary interest in one point of view and all the other experts are on the other side, there's a distinct whiff of bias out there, not indication of fundamental disagreement.

    I don't disagree about this, but we seem to have come to opposite conclusions about which side is which in this particular fight.

    @dkf said:

    BTW, if you've got any good ideas for technical fixes for systematically reducing CO2 levels in the atmosphere

    My fix is: DON'T. I think we're better off with more than we currently have, given everything we currently know.

    @dkf said:

    We can wait until we're certain, and then be certainly very sorry when we find we can't fix anything because we waited far too long.

    Dude, we're so far from that it's practically tragic.


  • Winner of the 2016 Presidential Election

    @boomzilla said:

    I don't disagree about this, but we seem to have come to opposite conclusions about which side is which in this particular fight.

    I'm pretty sure both sides are faking studies, because all parties involved have monetary interests, but I don't really care. Fossil fuels have no future anyway. Also, it's never a bad idea to try and become independent of limited natural resources, especially if the reserves are located in foreign countries with unstable/unpredictable governments.

    BTW: How many times have you guys had this exact conversation already?



  • @asdf said:

    Fossil fuels have no future anyway.

    TDEMSYR. Your life has no future, either. What's your definition of "future?"

    @asdf said:

    How many times have you guys had this exact conversation already?

    I don't recall having this one at all with @dkf.

    @asdf said:

    I'm pretty sure both sides are faking studies, because all parties involved have monetary interests, but I don't really care.

    It's not necessarily that people are faking studies, though they are, but that they aren't honest about obvious things. Either way, people making extraordinary claims (We're all gonna die!) have a burden of proof to meet that they haven't begun to touch.


  • Winner of the 2016 Presidential Election

    @boomzilla said:

    I don't recall having this one at all with @dkf.

    Sorry, I misremembered. A quick Google search on forums.thedailywtf.com returns 100s of "climate change" discussions, but it seems like @dkf never participated in any of those.

    @boomzilla said:

    TDEMSYR. Your life has no future, either. What's your definition of "future?"

    Why does this topic always trigger your Blakeyrant mode? :confused:

    Anyway, I don't think my great-grandchildren will live in a world where fossil fuels are as ubiquitous as they are now. I am convinced that in 50-60 years, we will finally be forced to seriously think about closed raw-material cycles for pretty much any resource we're using. I may be wrong about the timeframe, but this will have to happen eventually. The sooner we start to use our natural resources responsibly (especially valuable, non-recyclable raw materials like crude oil), the better our great-grandchildren will live.



  • @boomzilla said:

    extraordinary claims (We're all gonna die!)

    Was there a timeframe mentioned for this prediction? Just in a geological timescale, it seems like a safe enough bet...


Log in to reply
 

Looks like your connection to What the Daily WTF? was lost, please wait while we try to reconnect.