So while I don’t consider my premise an original thought, I don’t believe I’ve ever seen the silos displayed in a manor similar to the Rings of Knowledge (pictured below).

Note: I originally had one’s hard drive or mind as the center, but removed it after further consideration. Since that sort of data is not included in any other sources and can only be accessed via one method.

This allows you to visualize how each successive ring encapsulates all the data and knowledge contained within it.

It all started with email (Collaboration 1.0), then people got wise and started to save documents on shared drives. While both these solutions were a start, they were limited by connectivity, interfaces, and server space. The next step was the use of Content Management Systems (CMS). This allowed workflow, and web interfaces for accessing and searching your documents. CMS is still a great option for documents, but it was only one piece of the puzzle. We had documents, but the critical conversational pieces of information and thought processes were not accessible. We knew the result, but sometimes forgot how we arrived there.

To address this issue, independent wiki/blog/discussion technologies came on the scene. I am still a big fan of using wikis for development documentation as it seems to lend itself more to evolving with the development. I think discussions and blogs are sort of self explanatory as to what they accomplish. In total, these contribution driven technologies, sometime called web 2.0, give us another large piece of the puzzle. So how do we merge these all together?

Obi-Wan

This use of the StarWars quote by an former co-workers seemed like a great summation of my feelings when I heard that Oracle purchased Sun.  Course, this isn’t completely unfounded as I feel Oracle has sort of screwed the pooch with the acquisition of BEA a little over a year ago.  BEA went from having a strong innovative spirit focused on the middle-tier, to being forced to spend cycles merging their products with all the other random Oracle acquisitions.

Yet, being the optimist I am, I am hoping Oracle has learned from their mistakes and will actually use their new found power for good instead of evil.  So , here are a few of my thoughts on somethings Oracle could do that would be of value to me and hopefully Oracle.

Oracle Database Machines:  The highend database is one of the last little things to be virtualized, because of it’s need for optimal performance.  Oracle made it’s first investment in this with HP, with the Database Machine, but this cost a million dollars per machine.  I’d love to see Oracle use the Sun hardware division to create a reasonably priced pizza box db machine, potentially preconfiged and optimized.

This brings me to my second idea, which is creating application platform options at every tier with simple and well tested upgrade paths.  The example above could have an entry level machine with mysql installed, with an enterprise option built for Oracle grid installs using the power and technologies in the Oracle database.  They could do a similar option for the application tier, with an entry level development env using glassfish, and an enterprise version using WebLogic server.   The key for this to work is that the foundations for these tiers would need to be consistent, so you could upgrade easily, or select the right mix of technologies to support your needs.

We’ll see what Oracle does with it’s latest toys, I just hope that the open source movement doesn’t suffer from any poor oracle decisions.

My first challenge as a developer was getting in the mindset that publisher is simply a publishing engine that leverages templates. While much of the content seems dynamic when you are creating presentation templates using the PCS tags, it is really just running the data through the engine and creating html. My reason for attempting this is that I am trying to paginate the archive page of your standard news template in ALUI. This portlet template creates a main portlet, which shows the first X number of articles, and a link to an archive page that shows all the articles.

So after coming to terms with this eliminated your standard paging method of chunking data and going back to the source for data as needed, I came up with these approaches:

1. Dynamically create a series of pages that are chunked based on some arbitrary break point like items per page, or a time interval and create links to the different pages.
2. To pull back all the data and use JavaScript and dhtml to handle the displaying of the chunks.

I choose method number 2, since it seemed a little easier and the performance is better after the hit for the original load. This solution requires you to edit all three objects associated with your index.html file.

First, you’ll be adding a new integer property called maxitems to your Data Entry Template.


..
normal foreach loop over items
..
<pcs:if expr="(( item_position - 1 ) % maxitems ) == 0">
</div>
</pcs:if>
<pcs:if expr="(item_position == 1) &or; (((item_position - 1) % maxitems ) == 0)">
<div style="display:none" class="news" id="page<pcs:value expr="item_position"></pcs:value>">
<script language=javascript>
divcount++;
</script>
</pcs:if>
..
Your Normal Content Announcement Stuff Goes Here
..
After you close the foreach loop for items. Insert the following to create pagination links.
..
<pcs:if expr="items.length &gt; maxitems">
<pcs:foreach var="item" expr="filter(items, 'filtered.published &and; !filtered.hidden')">
<!-- <pcs:value expr="item_position"></pcs:value> -->
<pcs:if expr="(item_position == 1) &or; (((item_position - 1) % maxitems ) == 0) ">
<a href="javascript:showDiv(<pcs:value expr="item_position"></pcs:value>);">
<pcs:value expr="item_position"></pcs:value> - <pcs:value expr="item_position + 9"></pcs:value>
</a>
&nbsp;&nbsp;
</pcs:if>
</pcs:foreach>
</pcs:if>
<script language=javascript>
document.getElementById("page1").style.display="";
function showDiv(page){
for (x=1;x<=divcount*<pcs:value expr="maxitems"></pcs:value>;x+=<pcs:value expr="maxitems"></pcs:value>){
document.getElementById("page" + x).style.display="none";
}
document.getElementById("page" + page).style.display="";
}
</script>

After you publish your content item again, you should only see the first section of the page, and links to other sections. This is accomplished by creating hidden divs labeled page1, page1+maxitems, etc. When the page is rendered in a browser, the script at the bottom shows the first section, and updated the visible section when the JavaScript links are clicked.

I am in the process of updating the links to end at the last number and not a rounded amount based on max items. You can consider making this change as homework since it will show you understand how this actually works and some of the publisher nuances.

Eventually, I would like to combine to chunk up big sections onto different pages, and then use JavaScript to page through the smaller chunks on each page. This would limit the cost of the original load, while getting the flexibility and performance of JavaScript

So I was lucky enough to attend barcampphilly this past weekend which really got my mind racing about a ton of different topics.  One of the things that struck me though, was the adoption of twitter by barcampers.  It was cool, and a great way to communicate, but also a little scary.  So here are a few of my concerns with the social internet.

1) It’s not the quality of the site, but the adoption which is key.  This makes it important to be first to the market, and to drive adoption quickly.  The converse of this is that you need to build a much better mouse trap if you expect to incent users to switch.  I personally believe now that this concept exists it would be possible to build a better Twitter, but you’ll never get people to switch.  There is simply too much invested in this service and its connected apps.

2) It creates ADD (Attention Deficit Disorder)!  It was hard enough last year to get through a day of work when I had to respond to interruptions from email, IM, and phone.  Now it’s almost impossible to focus on a task without actually turning off distractions like twitterific.  Yes, I turn off twitterific during most of the day, as well as email and IM, otherwise I get nothing done.

3) Speaking of following hundreds, at what point does the stream of information coming from the RSS, Facebook and Twitter become drinking from the fire hose?  How does anyone filter all this information?  I am in a constant state with adding and refining my information streams.  Would anyone who is following more than 50 people on twitter let me know how you do it?

4) Visibility: not only friends, but employers, enemies, lawyers, etc all have access to your thoughts and actions.  The world (or followers at least) can see your what, where and why.  You are publishing your thoughts, feelings, eating habits, location almost everything there is to know about you.  Can you imagine what a consolidated view of all this data would look like?  Something someone said at barcampphilly this past weekend really struck me as amusing and true.  I’ll paraphrase, but it went something like this:

“Back in the 60s you would have screamed about the injustices of Big Brother, had you been forced to record you communication, location, eating habits, etc. Now people readily post this for the world to see.  It’s crazy.”

Oddly enough, many of these concerns create the benefits of social networking.  Its viral nature and ability to spread the word about a better product is incredible.  It’s astonishing how quickly a new YouTube video, blog post, or social app can gain notoriety.  I miss the simpler times when sites were slashdotted.  This same ability to spread at a grassroots level allows the common man to compete.  If you build a better site, service, whatever, you don’t need the mainstream media or investors. You just need to have adoption.

So while I continue my personal adoption of social networking and looking for better ways to use it, I also try and maintain my perspective and a healthy dose of caution that while I am sure this is going to change the landscape, it may not be the tool, technology or technique I am using today.

You can already tell this thing is going to be big when a flight from Philadelphia is impacted by it.  There are several Oracle shirts, people reading Oracle documents about development and pl/sql, iPhones, laptops, technology galore.  I personally enjoyed a few hours on the plane to knock off more of Good to Great, a book I’ve been trying to finish for awhile now.  It’s a great book and I highly recommend it, every page offers another insight.

Oddly enough, most of Good to Great contradicts the way Oracle has succeeded in becoming a technical powerhouse.  They have an extremely charismatic, egotistical leader, who seems to have expanded based upon acquisition.  I personally haven’t seen Oracle been great at anything since the database, and they seem to have actually lost market share in that which was once their core product.   I personally don’t care to much about the database product, and am mainly focused on their SOA suite.  I am hoping their latest purchase of BEA doesn’t go the way of some of the other acquisitions.  I am looking forward to seeing the path Oracle will set for its middleware suite, aptly called Fusion.  I have been to a few local sessions covering the “roadmap”, but still am not comfortable that the map they have provided is accurate as it’s a lot easier to say you are going to merge multiple mature products, it’s another to actually do it.

So what do I hope to come away with this week?  I hope to narrow down the products and technologies within the middleware stack that I can recommend we focus our effort, as unlike Oracle, I don’t think we can be a master of all.

At first I thought nothing of it, then I realized just how cool is the iPhone.  I had just received a software update that was essentially a complete replacement of the phone’s code.  I can get apps, mostly for free, that are developed by developers and friends trying to solve my problems, not by a corporation who doesn’t care about me and has slow release cycles.

So now not only is my iPhone the perfect device for my train rides with it’s ability to play podcasts, maps, music, videos, and obviously as a phone, but now I can do so much more.  I can stay current on twitter, post blog entries, find places to eat, it amazing what people think to build.

So thanks to everyone for their great work and for making the iPhone the cool phone around the water cooler.

One final note, here are my current applications of choice:

Labyrinth LE – just to show off the cool gyro features of the iPhone
Urbanspoon – since I can never decide on a place to eat when downtown
Air Sharing – so I can read some stuff on the train when I am not connected
VoiceNotes – seems to be more effective than Jott
Twitterrific – so I can tell people all about my exciting life (that’s sarcasm)

I should have tested it first, but I assumed that it would not convert urls to the tinyurl format since it was doing a straight post.  So I quickly started to rip apart the script and change it from calling curl to using a twitter.rb class I had found somewhere to do the post.  The intention was that it would be a lot easier to use ruby to parse the post and if required replace urls with a url from tinyurl.

Well after about 20 minutes I had tracked down the tweet.scpt as I had forgotten where I had installed it (~/Library/Application\ Support/Quicksilver/Actions/Tweet.scpt) and had updated it to call a ruby script and create a growl notification.  Then I search google for the tinyurl call and some ruby twitter stuff at which point I found the very cool twitter gem and this updated script http://tinyurl.com/2ds5kg which used the gem.

It required a few path fixes, but seemed very clean and had a few bonues…. I love having the twitter command line now,and it’s ability to have multiple users available for use (very cool).  So finally, I tested posting a url to see how it was working and where I needed to add my callout to tinyurl.  Well, damn if it didn’t make the conversion already.  I guess it must happen on the twiter side.  Oh well, it was still a fun experiment and provided some valuable insight into the capabilities of applescript, quicksilver, and ruby.

As you can see, while I am not a frequent blogger, it’s not a new technology to me, nor is IM or email. Yet, until recently I avoided the latest wave of technology sites and tools: the facebooks, twitters, myspaces, etc etc of the world. I am slightly uncomfortable with the web 2.0 unfocused public displays of personal “stuff”. It could be a self-confidence issue, but honestly I don’t think that’s the cause. I think it is more rooted in my uncertainty around its purpose and the limits of the web 2.0 universe.

At the conference, BDG created this great “social application” which borrowed many of its feature concepts from facebook, myspace and twitter. I found it very useful and addicting as I only imagine users of these web 2.0 applications do. The distinction is that there was a limit and a focus to this application…this conference was a group of somewhat like-minded individuals all in very close proximity. So when a question was posted about a topic, chances are I was interested regardless if it was about beers or the Plumtree portal. I even saw how these features might be of value in my professional life as an IT consultant. It would be great to know where my colleagues were so I could let them know if I knew someone there, or maybe had heard something of value about that client. Or wouldn’t this be such a better forum to ask a question instead of a mass email? And while I hate having 3 emails (work, home and spam) and 3 calendars; there needs to be limits. Everyone can’t know where you are going or what silly problem you can’t figure out, but I only want to post in one place, and read from one source.

It’ll be interesting to see where this all leads, corporate twitter servers like with IM, or just a more secure way to direct messages? I see it leading to a larger social community and more passive communication, it’s just a matter of where the universe ends?

Option 1: Buy an expensive framework like BEAs Liquid data which does most of what I need.

Option 2: Leverage and retool a portal framework, potentially even an open source one. To be basically an application with entitlements. I am personally not even sure how well this would work. It’s not a portlet I don’t want to display, but data in the portlet.

Option 3: This article seemed to have the right idea, but it leaves the presentation tear to the imagination. Maybe you could somehow have the objects represent themselves in XML(Yes, I hate xml too), and then using XSL as the display. Or if you have lots of time, you could somehow extend the bean tag libraries to handle the display of these role aware data objects.

I am obviously leaning towards option 3. I think the data filtering close to the source is a good idea, and with the filters in hibernate and other OR tools, probably not that complex to do. In particular if your application is not that complex.

Thanks in advance to anyone who offers some insight or advice.