My stakeholder would like users to have the ability to compare two PDFs side by side. They aren't content with users having to flick between PDFs (as normal). Instead, they'd like them to be embedded into a web page, next to each other. They would also like a drop-down menu above the second PDF to allow them to change the PDF to another one. I've included an image of this below to give you an idea of what this would look like. I hope you can see it:
I've built a quick and dirty demo of this working (see screenshot above) using iframes and a bit of javascript for the menu. The way this would work is that users would click on a button on a website saying 'compare PDFs' and that would bring up a separate screen in a new window which would contain only the two PDFs and the menu to switch PDFs (as per the image above).
The issue here is that although it's technically possible to achieve, it just seems so crude and wrong! But I need to be able to justify this with logic and reason. Is my instinct right, and if so:
what are the reasons why this approach is not recommended? (e.g. could it be damaging to our reputation if we put something like this on our website?! Why?)
do you have any alternative recommendations?
Or is this a perfectly good solution?!
Many thanks,
Katie
Katie, Your question is "why is this not recommended?"
Because computers are better at comparing data than people are.
This task is potentially tiring, inefficient and inaccurate.
It is essentially a screen-real-estate-management task, and doesn't add value by summarizing, interpreting, transforming, clarifying, innovating or enlightening.
It puts all the work on the shoulders of your users.
It violates the "Don't Make Me Think" principle.
This might be a benchmarking task, and you might not need every line of that data in those PDFs to deliver something innovative and valuable.
I challenge you to talk with your stakeholders and brainstorm innovative ways to combine or summarize the key data. Are these credit reports? Grades? Medical test scores? Phone bills? Legal contracts? There are lots of great patterns that you could use to pluck out a few values and compare them to each other. Think tables, graphs, or charts. Look at what TurboTax does with tax data after you are done filing your taxes -- this is what you paid, now look at what the average person with your income earned, deducted and paid. A simple table, with two columns, You and Everybody Else. What data are they looking at, and what will they learn by comparing the two documents side-by-side?
Summarize, enlighten, clarify, and remove noise. Cut to the heart of what they want to see, and just show them that.
Best to you!
Related
Having a DokuWiki with some content (regular to small in sice and depth) I would like to automatically generate a GraphViz or Freeplane or any Form of easy to grasp visualisation of my content.
Why? Because the wiki tends to become less and less effective, when searching and organizing its content. As a user I have no good way to get a sharp Idea of the Wiki structure, which is why more and more often topics are not written and found where they supposed to be.
How to generate graphical sitemap of large website is what I found so far, but because my wiki is not that big, it would be quicker for me to just manually make a graph. And because the main topics are not that often updated or extended (like 10 extension a month tops), it would not be that hard to keed it up to date manually.
However, I would like to avoid manual tasks, at least in the future.
So is there a plugin or any other good way to graph the contents?
starting on the landing page, following the internal-wiki-links
using the namespace-sitemap
Either one would be nice, 1. interest me a bit more, because it reflects the paths a user could go, when just calling the wiki-start-page. I am greatful for any help, thanks.
I wrote a simple tool to do just that, the graph can then be analyzed in Gephi. Have a look at this blogpost: http://www.splitbrain.org/blog/2010-08/02-graphing_dokuwiki_help_needed
First of all, I'd like to explain what I want to achieve.
Imagine that I own a leaflet drop business. I cover a city, which is divided into 10 sections.
I want to leaflet-drop those 10 sections systematically, and have a heatmap that shows when they were last done (ie, if an area was done just last week, it would show up as green, but if it was done 6 months ago, it would be red.)
However, occasionally, I would do additional leaflet drops within the area, that I wouldn't want to be included in the data above - these would then show up on the heatmap in, say, blue.
I've been looking at OpenLayers, and it seems like it can do what I want - but I'm not really that technical, so need advice. OpenLayers has the ability to select specific areas with a polygon tool, which is exactly what I need - but how can I input this data, which will change frequently?
I currently monitor coverage of the territories with an Excel spreadsheet, but would like this heatmap system in addition to that.
So - any ideas?
Heck, for a tool for you to use yourself, you could use a drawing package like Inkscape. Scan a map of the area, paste that in, draw polygons for the areas you care about, and then every time you do a drop, change the colour of the polygon.
OpenLayers is generally used by software developers to aggregate geospatial/map data from various sources into web applications used by many users. OpenLayers can do what you want but not without writing a fair amount of JavaScript code. You might be looking more for a tool like ESRI's ArcCatalog, although it is probably too powerful for your needs.
I did a quick google search of "best map drawing tools" and found SmartDraw, for example. I have no affiliation with SmartDraw, nor do I know if it's any good, but it seems like this type of software would be more suited to your needs.
For what you say would be more useful a desktop tool as e.g. http://www.qgis.org/ (is open source and free)
I'm using jQuery, ASP.NET, SQL Server, and the other usual suspects to design a company CRM. After they put in contact info, notes, dates, places and so forth they have to be able to select many different people to be "CC'ed." A group of people will be required to be one either "CC'ed" or "ToDo." The rest of the people can be nothing or "CC" or "ToDo." Currently we have it set up as a huge databind to templates with radio buttons for each option. Looks like shit. Anyone have any suggestions? I'd like to use a template with a datasource and have a good way to retrieve their answers and use them.
I'm leaning jQuery direction but like I said I'll need there to be up to 3 possible options for the people. This is going to be all opinion so I'm just looking for options.
Just to re-clarify, this concept is similar to email but I don't want them to have to type anything in as it is a set group of names that they're allowed to select from.
Looking for quick simple and pretty. somewhere in the range of 120 names.
If you intend to look down the jQuery route, I suggest that this widget could possibly help you out (even if only for inspiration sakes). http://quasipartikel.at/multiselect/
I'm struggling to "visualize" your form for terms of "real-estate expendature" etc.
Not directly what you are looking for, but this plugin may help
http://devgrow.com/slidernav-jquery-plugin/
Typing with intellisense. Sorry - any graphical thing will look overloaded.
Or: A table with filter options on top (again, typing). THere simply is no other way.
What I would most likely do to achieve this is implement the auto filter pattern that you type in a text box a few letters of the name and then it would filter down all of the overall results to those containing that pattern. Then have a select all button that will let you check all of them, and then the user can manually uncheck a few instead of having to check all.
The other thing to do would be to offer some type of categorization of the data so that they could filter by category that would put people in probable groups that would want all them all together. Like IT, HR, Executive or something similar.
Is tagging is the best user friendly way to categorize a subject? An example would be the tags mechanism used in this Q/A site. (StackOverflow.com). How you would Implement categories in a best user friendly way? Or hierarchical categories are the best user friendly way to present available categories?
Is there any online store use tag to categorize product categories?
What you want depends on the nature of the media being categorized.
If you're working primarily with media that is difficult to index, like images, audio, or video, then you want a loose tagging system that encourages putting as many tags as possible with each item. The more tags the merrier, because you will need to use the tags to help index the content for searching.
For something that is more easily indexed (text!), you want a much more rigid system where it may take an extra step or two to create new categories or even force users choose from pre-defined categories. You no longer need to rely on user-supplied tags for search indexing, because you can index the content directly. You are strictly doing categorization, and for categorization to have meaning you need to be sure that users are sorting things into the same categories.
Whether or not a hierarchy or tree structure is appropriate depends on how well your categories fit into a tree structure. Some things fit better than others, and many things that appear to fit a tree structure (like programming topics) turn out not to be such a good fit after all.
I say yes!
Tagging provides a many to many relationship between questions and categories. It makes them loosely coupled and so gives control as well as flexibility for categorizing things.
Tagging has the advantage that a single post can belong to more than one category. This is not possible with (conventional) hierarchical structures. Tagging is a more powerful system than hierarchical categorizing - and so it allows the users to express more.
A hierarchical structure works better if you have a single editor who will spend the time to neatly categorize everything so that a menu structure can be created.
I think tagging works well when things can/need not be strictly categorized or require a hierarchy. It's kind of a non-relational way of relating things.
If works, tags should be controlled or filtered (e.g. crystalreports vs crystal-reports).
Retiring duplicate tags & migrating subject to a proper category will require human intervention.
I don't know of any store that use tagging. Static attributes can be categorized (Electronics -> Camera -> SLR). There could be other attributes, which are dynamic (price range between x and y) & those are not tagged.
Amazon has tags as well. I think, its search option has one choice where it says "get products tagged with" & gives you a box to put in your idea of what a tag could be (which shows you a dropdown of tags, it finds matching - same as stackoverflow).
Tagging can be very useful in every case. For example, gmail applies tags to all emails. Tags could be predefined (with a fixed meaning), or user generated. Having those predefined tags is important as they work like categories, can be controlled by the store owners, and still don't force you to use a hierarchy. You can emulate a category system with tags by enforcing some specific rules on the type of allowable tags, but you can't go the other way round with hierarchical categories.
So, for example, to find all incoming emails on gmail, we can search by the inbox tag.
tag:inbox
Or to get all unread emails in the inbox, we can search by two tags:
tag:inbox && tag:unread
It's much better than trying to categorize in possibly this fashion, for example:
/Inbox
/Inbox/Unread
/Inbox/Spam
Not that anyone cares about spam (the actual ones), but spam can be unread too. So should we introduce another sub-category called Unread inside Spam, or do some other restructuring within Inbox itself?
/Inbox/Spam/Unread
Another store-specific example could be to find all SLR cameras on Amazon that uses compact flash for storage, and supports RAW, and JPEG formats, we could theoretically use:
tag:camera && tag:SLR && tag:Compact-Flash && tag:RAW-format && tag:JPEG-format
Categorization forces you to choose a hierarchy which might mean you have to pick one between two or more perfectly reasonable choices. For example, is an iPod Touch a music player, or a video player, or maybe a portable computer, or all of it?
its depend on the context, if you want to make structure clear and controllable you should force it by your pre-defined category, but if you want to make it flexible and make easier for your user tag is the best technique to manage subject, furthermore it makes search engine easier to analyse and provides better results
so my answer if you asking for user friendliness it would be tags.
We're currently working on a solution that involves managing a large number of parts for a project. In our database, we have a project table and we have a parts table. Those parts can be assigned to multiple projects and vise-versa. This is done through a link table.
We're happy on the database side and it wont be changed, however we're a bit stuck on how to display the UI in a user-friendly intuitive way.
There are about 6000 parts (...at the moment) and we need to be able to easily assign/unassign these parts from a project quickly and easily.
Does anyone have any good examples of this?
I have always found real-time filtering to be pleasant to work with and narrow down things. This can require your users to be somewhat computer literate though...
Furthermore I would consider something with D&D. I would imagine two lists (one with projects and one with modules) where I can multiselect on either list and drag a single item from the other list to that selection.
And if it is going to be an interface that's going to be heavily/repeatedly used, consider good keyboard support. Me personally find that repetetive tasks can be done MUCH faster if they can be accessed by hotkeys.
Just thinking out loud ...
There are two distinct parts to this. The first is selecting parts and projects from the database and the second is associating parts to/from projects. You should try to avoid doing both actions in a single dialog.
From your post, it appears that parts can be assigned to projects and that projects can be assigned to parts. So, a wizard approach might work:
project or part?
/ \
select project select part
| |
show list of parts show list of projects
| |
add/remove parts to project add/remove projects to part
When showing a list of projects/parts, use a simple, sorted list with a filter box. Using a tree or other categorised system can be ambiguous for the user. For example, if the list was of foods and there were categories for fruit and vegetable, where would you put tomato? A shopkeeper would probably put in in vegetables whereas a botanist would put it in with fruits. So, a simple sorted list with a search box (like FF's about:config) works surprisingly well.
I'd try a number of things, from a pure UI point of view, if your collection falls naturally into categories, I'm a big fan of cascading lists or a sortable, filterable grid. (or both combined)
Your choice will depend on your users computer literacy as well as space or technology constraints.
If you can spare the space, I think that two lists, perhaps selected items on the left, and the collection of items on the right with affordances such as checkmarks or >> << buttons are great.
I'd probably take a page out of Excel's book, it has some perfectly workable examples of this sort of thing.
I'd also take the time to add multi-level undo, working with large datasets is an absolute pain as a user when you have no recourse for simple mistakes, it should ideally track and handle whatever fiddly interaction your user needs to make.