Sometime in the past three weeks, the Iowa DOT document archive switched domains and got a new requirement:
http://www.mydotdocs.iowadot.gov/Home.aspx
QuoteNote: Must use Internet Explorer to download files from website.
"Smarter, Simpler, and Customer-Driven" my Mac-using ass. HOW can a state agency get away with implementing a restriction like this, especially in 2015, after IE user share has fallen for a decade? (BTW, I did try it on a PC, with Chrome and with Internet Explorer.
It didn't work.)
Does anyone have suggestions for how I may (figuratively and non-violently) yell at them in an e-mail to find out what's going on?
Government websites are not as user friendly, because they don't have to be, because the government, unlike private business, has nothing to lose by having a crappy site.
Looks like it's Microsoft end to end:
(https://www.aaroads.com/forum/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FFXlw83t.png%3F1&hash=ea4bc1ac3ce15266d3c882bc2f70ecf45deb74c5)
This kind of nonsense is nothing new. FEMA was rightly reamed out a few years ago for an online application (for individuals to obtain disaster relief) that required users to have the latest and greatest version of (IIRC) Internet Explorer. The problem in that case was the software was developed by a contractor, who used Microsoft development tools and couldn't be bothered to test the software with other browsers.
Doesn't even work with IE. I've shot them an E-mail myself asking why A) IE only and B) doesn't even work with IE. Included a screen capture similar to what Kurumi posted.
I have tried it out and I don't think the requirement to use IE is the issue. It works fine for me in Firefox until I click on the download link for a document. I think the code that confects the document download URLs has simply not been updated to generate URLs of the form that the server now expects to see before it offers a document for download.
Iowa DOT did implement user agent checking on its document archive about a year ago, with the result that wget wrapper scripts (I have several for it, two of which are designed to keep me up to date on signing plans) failed whenever wget revealed its true self to the server. I ended up adding a line of code to each script so that wget would false-flag itself as Firefox, and thereafter I had no problems. However, I haven't tried any wrapper scripts with the site in its current form. I assume they will collect download URLs successfully and then fall down when they attempt to download.
Edit (technical): The Iowa DOT document archive is fairly straightforward since all downloads are by HTTP GET request--no need to use the HTTP POST method and supply postdata confected in advance--but cookies have to be used.
Insult added to injury: I e-mailed the contact address on the website, records.center@dot.iowa.gov, and it bounced back as undeliverable. :confused: :banghead:
Why does downloading a simple PDF require both a Java applet and an ActiveX control? It seems that ActiveX is what is breaking compatibility with other browsers.
EDIT: The issue is that Windows is preventing the ActiveX control from running due to it being unable to verify the publisher (as it should, due to ActiveX's numerous vulnerabilities). It should only work if you have ActiveX put on its least restrictive setting, which is idiotic.
Plus, why would someone create an ActiveX control in this day an age period? Even Microsoft is abandoning the technology! (the removal of ActiveX is likely a major reason why IE12 was rebranded as "Edge")
Quote from: vdeane on May 11, 2015, 08:35:05 PM
Plus, why would someone create an ActiveX control in this day an age period? Even Microsoft is abandoning the technology! (the removal of ActiveX is likely a major reason why IE12 was rebranded as "Edge")
For real. With HTML5, there is practically nothing you can't do in native HTML. Even Flash is on its deathbed. Creating a browser application in 2015 that requires a third-party plugin shows a woeful lack of understanding about current Web standards.
IMHO, current web standards are a mess.
"HTML5" has been hyped since the end of the 1990s yet all the specifications are still not fully ratified. Compatibility for HTML5 and CSS3 is still all over the place when comparing various web browsers on different viewing platforms. HTML5 and CSS3 work best by far on traditional desktop & notebook computers. Compatibility drops off dramatically on various smart phone and tablet models. Game consoles and smart TV sets are worse. Internet Explorer 8 on aging Windows XP PCs was a drag on HTML5 and CSS3, but IE8 was far from being the only problem. The situation has been improving in the last couple years, but not enough.
Regarding Flash, yes it's on its deathbed in the mobile device space. But it's still alive and well on desktop computers. The difference between Flash and HTML5-based features is more of a philosophical difference than anything else (Adobe being the technology developer rather than it being open source). Flash may be a plug-in, but the code for HTML5 features like SVG graphics, Canvas, CSS3 functions and various audio/video codecs might as well be plug-ins. You just get them already embedded in the browser instead of having to manually install them. Google Chrome builds in its own Flash and PDF viewers by the way.
Performance wise, there's hardly any difference between Flash and HTML5. I can encode video or vector-based animations in either standard that will either work well or bog down a computer into choppy hell exactly the same way.
The best looking web sites have a lot more under the hood than straight HTML and CSS. Someone who lives, eats and breathes computer programming might be able to keep up with that. I think the growing complexity has driven a lot of individuals and small businesses away from maintaining pure web pages. They've gone to the walled gardens of Facebook, Twitter, Pinterest, etc. or if they're still making web pages they're using canned templates like those from WordPress. It's not easy to design a great looking and properly functional web site from scratch anymore, at least not if you're going to hand code it yourself.
Youtube's and Google Maps's HTML5 implementations are much memory-hungrier than Flash was.
Quote from: Bobby5280 on May 19, 2015, 04:52:27 PM
IMHO, current web standards are a mess.
"HTML5" has been hyped since the end of the 1990s yet all the specifications are still not fully ratified. Compatibility for HTML5 and CSS3 is still all over the place when comparing various web browsers on different viewing platforms. HTML5 and CSS3 work best by far on traditional desktop & notebook computers. Compatibility drops off dramatically on various smart phone and tablet models. Game consoles and smart TV sets are worse. Internet Explorer 8 on aging Windows XP PCs was a drag on HTML5 and CSS3, but IE8 was far from being the only problem. The situation has been improving in the last couple years, but not enough.
Regarding Flash, yes it's on its deathbed in the mobile device space. But it's still alive and well on desktop computers. The difference between Flash and HTML5-based features is more of a philosophical difference than anything else (Adobe being the technology developer rather than it being open source). Flash may be a plug-in, but the code for HTML5 features like SVG graphics, Canvas, CSS3 functions and various audio/video codecs might as well be plug-ins. You just get them already embedded in the browser instead of having to manually install them. Google Chrome builds in its own Flash and PDF viewers by the way.
Performance wise, there's hardly any difference between Flash and HTML5. I can encode video or vector-based animations in either standard that will either work well or bog down a computer into choppy hell exactly the same way.
The best looking web sites have a lot more under the hood than straight HTML and CSS. Someone who lives, eats and breathes computer programming might be able to keep up with that. I think the growing complexity has driven a lot of individuals and small businesses away from maintaining pure web pages. They've gone to the walled gardens of Facebook, Twitter, Pinterest, etc. or if they're still making web pages they're using canned templates like those from WordPress. It's not easy to design a great looking and properly functional web site from scratch anymore, at least not if you're going to hand code it yourself.
Actually, that's not true... html5 has, in fact, been ratified: http://www.infoworld.com/article/2839557/application-development/html5-finally-reaches-official-status.html
It seems like web sites themselves are a dying medium on mobile devices, with the modern focus on "apps", to the point where many new services are inaccessible if you don't have a smartphone or tablet.
Ratified and fully ratified are two different things. Even if the W3C approved all of the different technologies included under the HTML5 umbrella there is still a lot of work to do with compatibility. Web browsers don't all have the same level of compatibility with HTML5. The differences get worse on mobile devices, smart TV sets and game consoles.
If you code together a great looking web site using all the latest bells and whistles of HTML5 and CSS3 those features will work only on a few of the newest web browsers. They'll be broken on everything else. At least with Flash it worked on everything that supported Flash. And it didn't take some international committee many years to agree on something to make Flash work.
Regarding apps, I think people are badly overestimating their value. I am very picky about what apps I install on my smart phone. I flatly refuse to install anything that is nothing more than a commercial for some business. Web sites as apps are a total waste of valuable, finite memory. A lot of Android phones have only 2GB of space available for applications. Apps like Facebook consume more than 100MB of that space, especially when you factor in the updates.
Is "fully ratified" even an official thing or is that just something you came up with? Last I checked, the W3C can't make anyone adopt anything. They just set a standard, and it's in the browser maker's interest to follow it so that a bunch of websites don't break on their browser.
Has anyone actually confirmed that the Iowa DOT EDMS will work if IE is used with a Java applet and ActiveX controls allowed? Has anyone actually had Iowa DOT explain why they took this huge step backward?
Quote from: J N Winkler on May 22, 2015, 10:50:26 AM
Has anyone actually confirmed that the Iowa DOT EDMS will work if IE is used with a Java applet and ActiveX controls allowed? Has anyone actually had Iowa DOT explain why they took this huge step backward?
I can get the applet to load (they use the Daeja ViewONE applet) but can't view any of the pages in the documents. I tried copying the "Download Original File" link into the "Open URL" menu, but it doesn't work.
Today, I discovered that two separate e-mails from the DOT had been flagged as spam and sent directly to the trash. I don't know why: It was either the image attached as a signature, the image used as a background, or the fact the whole thing was written in Comic Sans.
They'll get something from me by Monday.
Quote from: Jeff Morrison on June 14, 2015, 01:30:32 AM
Today, I discovered that two separate e-mails from the DOT had been flagged as spam and sent directly to the trash. I don't know why: It was either the image attached as a signature, the image used as a background, or the fact the whole thing was written in Comic Sans.
They'll get something from me by Monday.
This is not making me confident about their computer skills... :wow:
I have now successfully downloaded from the site, without having to use IE. They attempted to re-skin the existing EDMS but their code is buggy, so the download URLs are incorrect.
An example:
Incorrect URL:
http://www.mydotdocs.iowadot.gov/UserControls/GetFile.ashx?DocId=3808041&PageId=17579982&FileSequence=1&Filename=20404433.pdf&PageNumber=0&DisplayType=R&Size=O&Type=O
Correct URL:
http://www.mydotdocs.iowadot.gov/GetFile.ashx?DocId=3808041&PageId=17579982&FileSequence=1&Filename=20404433.pdf&PageNumber=0&DisplayType=R&Size=O&Type=O
String "UserControls/" replaces "CMEPortal/" in URLs supplied by the old version of the Web interface. For current URLs it appears to be redundant and, of course, breaks downloading with 404 errors. (I don't think either link above is necessarily directly clickable from this forum, unless you happen to have an existing session with the EDMS in the same browser window. I don't know about the current version of the Web interface, but the old one required a cookie to download that was cut only when you passed through the entry page.)
There is no apparent need to use IE other than to use their worthless viewing applet.
Quote from: J N Winkler on June 15, 2015, 03:19:39 PM
String "UserControls/" replaces "CMEPortal/" in URLs supplied by the old version of the Web interface. For current URLs it appears to be redundant and, of course, breaks downloading with 404 errors. (I don't think either link above is necessarily directly clickable from this forum, unless you happen to have an existing session with the EDMS in the same browser window. I don't know about the current version of the Web interface, but the old one required a cookie to download that was cut only when you passed through the entry page.)
Awesome. :clap: Great to know there's a workaround by taking the "UserControls/" out of the URL. (Or, for those who actually want to use this viewer applet, they can copy the URL of the "Download Original File" link, right-click inside the applet box, select Open URL, paste the URL, take out the "UserControls/" in the URL, and click OK to view the document.)
You are welcome. I am just happy we have established that the data is still available online. I'm in the process of rewriting my EDMS downloader script so that it will work with the new site, preferably in a way that won't break when they finally get around to fixing the URL problem.
After two months of silence, I asked what was up and got a very short e-mail. What is the nice way of saying "Requiring Internet Explorer DOES NOT mean it's working, it means it's STILL BROKEN!"
QuoteThe myDOT Docs site is up and working again, but you do need to access it with Internet Explorer, none of the others will work.
If the workaround above works, fine, but holy crap is that a time-consuming endeavor when repeated over and over for every page.
Quote from: Jeff Morrison on August 11, 2015, 01:57:38 PMAfter two months of silence, I asked what was up and got a very short e-mail. What is the nice way of saying "Requiring Internet Explorer DOES NOT mean it's working, it means it's STILL BROKEN!"
QuoteThe myDOT Docs site is up and working again, but you do need to access it with Internet Explorer, none of the others will work.
If the workaround above works, fine, but holy crap is that a time-consuming endeavor when repeated over and over for every page.
The workaround does still work; I run my Iowa DOT EDMS downloader on a monthly schedule and it pulled in 45.1 MB of signing plans just last night.
You can test it anytime by first opening the MyDocs site in a separate browser tab and going in through the front door ("Accept terms of use," etc.) to cut a cookie, and then clicking on the "Correct URL" in my post upthread. If the workaround still works, then your browser should ask you if you want to save a PDF file locally or open it in
Acrobat (or whatever Windows has designated as your default viewer for PDF files).
It seems your correspondents at Iowa DOT are hung up on Internet Explorer being necessary to view documents in a Web browser. The real problem is that the EDMS is supplying incorrect URLs due to a coding bug. We know the URLs are incorrect for downloading. They may also be incorrect for viewing if whatever IE-only Java applet/ActiveX control they are using takes document URLs as a starting point.
I strongly suspect the lack of support for multiple browsers and the requirement to use a special Java/ActiveX plugin for viewing is to accommodate the fact that the documents are supplied from an EDMS that cannot offer random-offset read access to the files it contains. Until this limitation is addressed, I think it is better for end users such as ourselves to push for a download mechanism that actually works, rather than the ability to view the documents online in a browser of our choice.
In order for downloading to work without the need for the end user to hand-edit every download URL, the coding bug that generates incorrect URLs must be fixed.
If you continue to correspond with Iowa DOT about this issue, my strong recommendation is that you drop all discussion of the viewing side of things, and stress that there is at least one coding bug that is generating incorrect download URLs and must be fixed. I'd also send them a link to this thread since it explains what the bug is (though the phrase "worthless viewing applet" won't win friends with them).
Oh, this is just lovely.
QuoteAn error occurred while connecting to the Portal as guest.
Now I can't even get into the site to try the hand-code editing method.
I think this is strictly a short-lived error--if you tried it shortly before you posted, it may have been because the site was taken down for a midnight update. I tried the portal just now (in Firefox!) and it worked for me.
Bumping this thread... it looks like they've dropped the IE-only restriction. I was able to access documents via Edge without issue.
Edge is just a rebranding of IE.
Quote from: vdeane on May 10, 2016, 01:28:29 PM
Edge is just a rebranding of IE.
True, but I tried it on Chrome and it worked as well.