|
|
As of yesterday, the readership on this AI3 blog passed 3000 daily for the first time. It has been steadily inching upward, and finally passed that minor milestone. Thank you!
I’ve been writing this blog for five years now, with some 400 total posts, or about 1.5 blog posts per week. I know my style is toward longer articles and less frequent posting, most often of a fairly detailed or technical nature. And, while I have a Twitter account, I do not bleat. My style is for more meaty discussions. Perhaps it belies my age. 
The real growth in this blog, however, has come about with my conscious attempt to write for the enterprise audience. RDF, the semantic enterprise, linked data and ontologies need a bridge from the technical community to the one of practitioners. Much progress and uptake has been occurring with these business and government audiences.
At the recent SemTech meeting, I was taken aside by many individuals noting my blog posts and thanking me for the thought and effort behind them. Thank you for noticing, and reading, and you are welcome. We need more translation of semantic topics and technologies to pragmatic terms.
If you have been following the standard W3C and SemWeb mailing lists recently, you will have noticed an anxiety and a continuation of the fractious nature of this “community”. In part this comes about because there are efforts afoot to revisit the RDF specs. But, mostly, I think, it is the ongoing nature of many in this group to snatch defeat from the jaws of victory. The search by some for perfection and insistence on parochial needs and preferences can give a pettiness to this “community” that is unbecoming.
Many of us have abandoned those forums for those reasons. As for myself, I will continue to evangelize to the buying market and keep the gaze pointing outward. There is a wealth of need for tools, techniques, methods, documentation, structures, and narratives. Thanks to all of you, the readership of this blog, for continuing to affirm this value.
So, in the great scheme of things, the readership of this blog is quite small in comparison with the big boys. On the other hand, very few individuals have higher numbers, and all of this for a fairly esoteric area. I think this proves there is a market and a need out there for semantic solutions.
Thanks again! And, for those in the United States, have a most enjoyable 4th of July holiday!
OK. After an experiment of more than three years, I have just now canceled my Google AdSense participation. (Which, Google, by the way, makes almost impossible to do: Finding the cancel link is hard enough; but who remembers the day they first signed up for ads and how many impressions they got that day? Both are required to get a cancellation request approved. Give me a break. It is worse than banks claiming small digits from bank interest for their own income!!)
Despite my sub-title, I never did expect to make much (or, really, any) money from Google ads. When I first signed up for it in Dec 2006, I stated I was doing it to find out how this ad-based business really works.
Well, from my standpoint, it does not work well; actually, not well at all.
Over the years I have seen visits on this site climb to nearly 3 K per day, and other nice growth factors. Perhaps if I were really focused on ad revenue I would have rotated stuff, tried alternative placements, yada yada. But, mostly, I was just trying to see who made out in this ad game.
It is certainly not the standard blog. I think my stats put me somewhere in the top 1% of all sites visited, but even that is not enough to even pay my monthly server charges (now higher with Amazon EC2).
Yet, in recent months, I have noticed some vendors have specifically targeted advertising on my blog and there also has been an increase in full banner ads (away from the standard, unobtrusive link Google ads of years past). Maybe they know something I don’t and they are winning, but my monthly ad income has dropped or remained flat.
And, then, I began to get full panel flashing ads on my site that just screamed Hit me! Hit me!. WTF. It was the last straw. Where did the unobtrusive link stuff go? Screw it; I can afford to pay my own monthly chump change.
This is probably not the time or place to discuss business models on the Web, but the woeful state of ad-based revenue is apparent. My goodness, I’m getting tired of ReadWriteWeb, as an example and one of the biggest at that, shilling with repeats and big ads with stories for their prominent advertisers each weekend. And, they are one of the only few ad winners!
My honest guess is that fewer than 1/10 of 1% of Web sites with advertising make enough to cover their bandwidth and server costs. How do you spell s-m-a-r-t?
So, the experiment is over. I will now think a bit about how I can reclaim that valuable Web page space from my former charitable contribution to the Google cafeteria. Bring on the sushi!
As we see more collaboration forums emerge, one question that naturally arises is the joint authoring or editing of images. This is particularly important as “official” slide decks or presentations come to the fore.
There are perhaps many different ways to skin this cat. In this article, I describe how to do so using the free, open source SVG editing program, Inkscape.
Like many of you, I have been creating and editing images for years. I am by no means a graphics artist, but images and diagrams have been essential for communicating my work.
Until a few years back, I was totally a bitmap man. I used Paint Shop Pro (bought by Corel in 2004 and getting long in the tooth) and did a lot of copying and pasting.
I switched to Inkscape about two years ago for the following reasons:
Once you have a working image in Inkscape, make sure all collaborators have a copy of the software. Then:
]; this will resize and put your target image in the full display windowOf course, it is more often the case that not all collaborators may have a copy of Inkscape or that the image began in the SVG format.
The image below began as a Windows Powerpoint clip art file, which has then gone through some modifications. Note the bearded guy’s hand holding the paper is out of registry (because I screwed up in earlier editing, but I also can easily fix because it is a vector image! ). Also note we have the border from Inkscape as suggested above. This file, BTW, is people.png, and was created as a PNG after a screen capture from Inkscape:
When beginning in Powerpoint or as clip art, files in the format of Windows metafile (*.wmf) or extended WMF (*.emf) work well. (For example, you can download and play with the native Inkscape format of people.svg, or the people.wmf or people.emf versions of the image above.) If you already have images in a Powerpoint presentation, save in one of these two formats, with (*.emf) preferred. (EMF is generally better for text.)
You can open or load these files directly into Inkscape. Generally, they will come in as a group of vectors; to edit the pieces, you should “ungroup.”
After editing per the instructions in the previous section, if you need to re-insert back into Powerpoint, please use the *.emf format (and make sure you do not save text as paths).
For example, see the following PNG graphic taken from a Inkscape file (figure_text.svg):
We can save it as an EMF (figure_textpath.emf) to a Powerpoint, with the option of converting text to paths:
Or, we can save it as an EMF (figure_text.emf) to a Powerpoint, only this time not converting text to paths and then “ungrouping” once in Powerpoint:
Note the latter option, text not as path, is the far superior one. However, also note that borders are added to the figures and vertical text is rotated 90o back to horizontal. Nonetheless, the figure is fully editable, including text. Also, if the original Inkscape figures are constructed with lines of the same color as fills, the border conversion also works well.
Frankly, especially with text, because there can be orientation and other changes going from Inkscape to Powerpoint, I recommend using Inkscape and its native SVG for all early modifications and to keep a canonical copy of your images. Then, prior to completion of the deck, save as EMF for import into Powerpoint and then clean up. If changes later need to be made to the graphic, I recommend doing so in Inkscape and then re-importing.
I should note there is an option, as well, in Inkscape to convert raster images to vector ones (use Path -> Trace bitmap … and invoke the multiple scans with colors). This is doable, but involves quite a bit of image copying, manipulation and color separation to achieve workable results. You may want to see further Inkscape’s documentation on tracing, or more fully this reference dealing with color.
Of course, there are likely many other ways to approach these issues of collaboration and sharing. I will leave it to others to suggest and explain those options.
My wife and I are not gamblers, and were somewhat surprised to find ourselves at our local destination casino last weekend to see a concert by Boz Scaggs and to spend the night in a high-roller suite with a glassed-in shower and electrically controlled window shades. The only missing piece was a mirror on the ceiling. Of course there was an occasion involved, and from top to bottom we had an absolute, total great time.
The highlight of the whole affair was Boz Scaggs himself and his band. Boz Scaggs goes back to our courtship; and we celebrated our 30th wedding anniversary this year! But, this was not a geriatric trip down memory lane: this was top-drawer, great music and entertainment. Not to get too excessive, but this show was close to one of the best I have ever seen!
I normally would not comment on such matters on this blog. After all, I’m generally trucking down an esoteric trail with an audience that at most fills living rooms, not concert halls (let alone stadiums). Such is the semantic Web today.
But then one of those somethings happened this week: I was asked to go down memory lane and resurrect some of my older posts. I read quite a few from years back, and liked some of what I read. It was actually kinda fun. And, I had forgotten many of these older hobby horses or even that I had written them.
Now, in the original Stone Age days of this blog, namely 4-5 years ago, I had like 20 – 30 readers per day. Today, I’m closer to 2500 per day, and seemingly growing pretty steadily. I also now have a backlog of about 400 prior posts. Most have not been read, or at least not by any notable readership.
So, like Boz Skaggs, I decided I would on occasion bring back one of those older contributions that maybe did not get too much airplay in the older days. And, since these are re-treads, I should also re-introduce them on Friday when the news cycle is slow and no one is really very attentive anyway. I mean, afer all, they are only electrons!
So, with this convolution, I’m pleased to introduce this occasional Friday re-release of selected earlier posts. I may make some minor changes to these older posts to make them current or correct typos and such. If I do, I will so note.
I do not have enough historical backlog of posts to warrant a re-tread every Friday. But, on occasion, including this Friday, I will post again. Look for the brown bag symbol on these reprised posts.
(Holy Leap Year, Batman!)
I’ve stated many times I hate WordPress upgrades. I know the sponsors have tried to make it easier over time, but upgrades are still painful, wrought with risk and error, and always force me to research and figure out what went wrong.
I last upgraded to WP v. 2.2.1, and with a real rant to accompany it.
Since then, some of us had been seeing some insidious stuff getting inserted into our RSS feeds, but had not been able to stem it. Then, I was doing my normal morning systems check and saw that my site was completely down, completely blank. Grrrr. Who knows what that specific problem was.
Version 2.3.3. had been announced with a fix for the RSS feed spam problem, so, rather than trying to diagnose and fix my current version, it was time to upgrade. (Grrrr.)
But, then I realized, possibly by doing so, I might also see a fix to a longstanding issue I had had with plug-ins somehow limiting my chronological listing of past posts. (Hooray!) That one had really been sticking in my craw, had caused me to de-activate some plug-ins I thought useful, and had led to only a handful of prior posts appearing.
So, the upgrade was made. Sadly, no problems (other than the XML-RPC implementation issue) were solved. And, unfortunately, my chronological listings still only displayed when throttled back to the past 30 or so. (Grrrr.)
Well, s**t. So after (for what was for me, with some of my more complicated site aspects) nearly a two hour minor upgrade, the only real benefit I or my readers would see is that the site was no longer blank! This hardly looked like a good deal.
So, assuming the chronology problem fix was not near at hand, I decided to manually add the past entries back to my chronology page. (Actually, this sounds worse than it really is since I have learned some quick tricks for gleaning listings from other sites; I just turned those techniques on my own blog!). While grinding teeth to nubs, I did what everyone who works intimately with software often does: I did the workaround.
So, now all full listings have been restored (though still with some recent postings overlap; Grrrr).
What brought a smile was seeing some posts from a year or two ago that I liked and had completely forgotten; some others brought a shudder. Here are some older personal favorites:
Nonetheless, now all 250 or so posts on my site from Day 1 in early 2005 can now be seen again; it has been awhile! 
Naturally, that was not the end of the saga.
After making the upgrade, I noticed that all category listings and lookups had been wiped off my blog. I could see them in the MySQL and the editor still had the listing, but the site itself and the admin panel were blank.
Grrrr. (Try to stay calm and not panic.)
It’s another one of those deals where it is time to search like crazy and hope that someone more knowledgable than me has encountered the same problem and fixed it. Sure enough, in an obscure reference, I got the glimmer that maybe re-starting MySQL could fix the problem.
Well, it did. But go figure. . . .
Thankfully, my Advanced TinyMCE plug-in that gives more editing functions works great for me in WP v. 2.3.3. At least that is a relief!
And so, we end on an anti-Grrrr note. Sweet dreams.
The issue of popups, thumbnails, link indicators, and other visual clues for blog content has been an interesting and difficult one. When Snap first came out with its preview popup thumbnails of referenced links (”Snap Shots“), it became all the rage until there was a backlash against ‘popupitis‘.
Similarly, many of us, for styling and design considerations (perhaps not always for the best?!), have mucked around with our CSS to the point that a standard link is sometimes hard to discern. You’ve seen them, and I have myself been guilty:
As we get clever on this, we then need to compensate with other visual clues for the link.
In my case, about a year ago I adopted the terrific Link Indication WordPress plug-in by Michael Woehrer, which enabled me to type-by-icon the kind of link you, the reader, sees. In my own case, I had icons (for example) for Wikipedia, PDFs, RDF, general external links and some others. The idea, of course, is that faithful readers would learn these subtle distinctions and appreciate the visual cues. (Now for the obligatory, yeah, right!)
To avert symptoms similar to popupitis, it is important to keep these visual cues subtle and (hopefully) unobtrusive. I was actually fairly proud of my Link Indication icons in this regard.
I then began playing with zLinks about two weeks ago, and wrote a blog posting about it. Check that out and the update blog notice from Fred Giasson to learn more. And, if you have WordPress, you can download and install the plug-in yourself.
But now the game has changed. Instantaneously, my links became more meaningful, and my link representations on my blog more fat.
The links became more meaningful because now I had the wealth of linkages and relationships tied to every single embedded link on my writings. I have been an aggressive “linker” and this has meant a hidden wealth of interlinkages automatically available to my postings and writings. Sure, I don’t often or always want to explore this richness (and, maybe, many if not most of my readers don’t have that interest all the time as well), but, simply having it there has opened my eyes to what has been called ‘linked data.’
Further, the basis of relating a link to a MIME type or similar document-level distinction now seems primitive. The meaningful distinction is no longer whether the document is a Powerpoint or PDF, but what subjects it is about and who, what, where and when it describes. The link now becomes not a doorway to a document house, but a reference to individual rooms or objects therein.
This richness and its implications are only now becoming apparent to me (and in a still-forming way). Moreover, through such things as backlinks, directed connections, implied connections and many others, this now-emerging world of interconnectedness is still revealing itself.
The new branding of the Zitgist Browser Linker to zLinks, I think, is a nice acknowledgement by the developers that something fundamentally new is afoot. It has been exciting (and rewarding to me) that as one of the early users of this capability that the developers (Fred, especially, thanks!) have sought me out for input and ideas.
The enhancements in this most recent Zitgist release tell me we have truly entered the era of the ‘Power of Z.’ Namely, the reach of a zLinks link is to make real today’s basis to deliver data interconnectedness. This is not the future; it is today. And, it is profound and exciting.
So, with a breaking of document classification boundaries (such as MIME type) to one that is now attuned to atomic data, any imaginable classification scheme becomes possible. But in this open typing, how do we handle the poor, overburdened link? How do we convey its power and reach? We’d like to convey some meaning, but where does it end? Readability would never accept Dewey Decimal tags or literal metadata text or any other such construct appended to the standard link.
From a practical standpoint, my first challenge was including the standard zLinks “mini-Z” icon associated with the zLinks popup that is the entree point to all of this interlinkedness richness. (By the way, have you been mousing over these icons to see the cool zLinks popups? Let alone following those reference links to their own Zitgist template reports?) The problem was, here was another new and diverting icon on top of the ones I was using with Link Indication — in other words, my link representations were becoming fat.
To add insult to injury, when I, as blog author, need to annotate or make other local notes on my local zLinks capabilities, I also need to call up and deal with the zLinks annotation facility. And, it too, has its own icon. So, after installing zLinks, I found I was now suffering from a new disease, iconitis, that has symptoms dangerously close to popupitis.
Thus, here is what one of my links looked like with the standard Link Indication icon and the zlinks annotation and standard icons while in authoring mode:
My gawd, my links were getting as adorned with all manner of fruits and nuts worse than tutti frutti.
Since I am as much in authoring mode as not, this distraction is in my face about half of the time. So, my decision: Get ‘link lean’ — skinny down those link icons and references, sufficient to where things again become usable and readable.
It was time to say goodbye to Link Indication.
There is really no need to make a heavy point of this except to note that the Web will continue to be ubiquitous as an access point to information, that information will devolve to be object- and data-centric and not at the document level, and the link (in keeping with its essence of the Web) will be the essential gateway for access.
I like the decisions Zitgist has made for zLinks: to provide a single, subtle and small icon, that itself brings up its own dialog showing the richness of the linked data support behind the embedded link. This popup is made available only when desired after a mouseover with a short delay (keeping the popup hidden during standard mouse movements). But then, when invoked, a new separate world of data types and links with expandable icons and tooltips is revealed:
This richness can be shown in the following example zLinks popup for the embedded link to Sweet Tools, in which all 600 tools are made available from a single link! This scrollable and extensible design is very much in keeping with growth and potential and meaning for the once lowly link:
So, with zLinks, I and my readers may have now given up showing links by MIME type, but we have gained the power of complete connectedness with the Web.
Let’s all raise a toast to the ‘Power of Z’ and to keeping links lean!
According to the WordPress codex, my Advanced TinyMCE Editor has been tested and found compatible with WordPress v. 2.3. (About 160 total plug-ins have been so verified.)
This new WP version adds enhanced support for tagging, among others (see Aaron Brazell’s 10 things you should know about this release).
I will be doing my own testing when a stable release is issued, but this is good news for this popular plug-in. You may get version 0.5 of Advanced TinyMCE from here.
Zitgist’s Plug-in Exposes Linked Data for Hundreds of Thousands of WordPress Sites
)The essence of the Web is the link. We use it to navigate, discover, form communities and get high rankings (or not!) for our Web pages on search engines. But, each link carries much more behind it than what has generally been exposed. That is, until now . . . .
Frédérick Giasson is a pragmatic innovator of the structured Web and semantic Web. Most recently, his efforts have included Ping the Semantic Web (that aggregates RDF published on the Web), the Zitgist semantic Web browser (that enables that RDF data to be viewed in useful ways), TalkDigger (for finding and sharing topical Web discussions), and efforts on a variety of ontologies, including jointly with me on UMBEL.
I have been an aggressive “linker” for some time and try to refer to Wikipedia often for definitions or background as well. Thus, Fred’s most recent efforts to continue to add value to the link as the basic coin of the Web realm really caught my eye.
In the early days of the Web, links were used solely to visit specific Web pages or locations within those documents. Somewhat later, actions such as searching or purchasing items could be associated with a link. Most recently, with the emergence of the semantic Web, the very nature of the link has become ambiguous, potentially representing any of the link’s former uses or either direct or indirect references to data and resources.
The Zitgist zLinks plug-in now makes these link uses explicit from within WordPress blogs.
Thus, we see that links can fulfill three different purposes, in rough order of their emergence:
The emergence of linked data and the semantic Web (or at least the provision of data via the structured Web) are making the use of the link more complicated and ambiguous. Moreover, sometimes a link is an indirect reference to where data exists, and not the actual resource itself.
What Zitgist’s zLinks does is to make these uses explicit and to remove ambiguities. Further, if a link is not to an actual resource but only a reference to it, zLinks resolves to the link’s correct destination. And, still further, a zLinks link is the gateway to still additional links from its reference destination, making the service a powerful jumping off point in the true spirit of the interlinked Web.
To my knowledge, zLinks is is the first and purest implementation of what Kingsley Idehen has termed the “enhanced anchor” or <a++>. RDFa and embedded RDF have similar objectives but are not premised on resolving the existing link.
Like the SIOC Import Plug-in, which imports SIOC metadata into a WordPress blog, the zLinks tool recognizes the importance of standard blogging software and automated background tools to expose data and capabilities. Since WordPress has many hundreds of thousands of site owners and bloggers — not to mention hundreds of millions of visitors — zLinks could be an important first exposure for many to the real power of linking and the semantic Web.
As a site owner, zLinks works identically to other plug-ins: simply install it and then it works smoothly and easily.
As a site user who might encounter a zLinks icon in a WordPress blog, all you need to do is click on mouse over the zLinks launcher icon at the end of any visible link. You will first get an alert that the system is working, retrieving all of the necessary background link information. You will then get a popup showing the results, similar to this one for my own AI3 blog:
The zLinks popup offers direct and related links, with the icons and other associated information an indicator as to the nature of the link and its purpose. In our example case, I click on my name reference, which brings up my FOAF file in the Zitgist browser:
Note how picture, mapping and other information is automatically “meshed” with my FOAF file. From this Zitgist browser location, I could obviously continue to explore still further links and relationships. In this manner, zLinks adds an entirely new dynamic dimension to the concept of ‘interlinking.’
If the initial zLinks link references data, that data is now resolved to its proper direct location, and is presented as RDF with further meshing and manipulation available. Other resources may take you directly to a Web page or perform other actions. Some of those actions, for example, may be to format data results in specific views (timelines, maps, charts, tables, graphs, structured reports, etc.). If the sources are data, the ability to make transformations or present the data in various views opens a rich horizon of options.
I made some minor tweaks to the Zitgist distribution as provided. First, I replaced the initial link icon — 
– with this one –
– that is smaller and more in keeping with my local WordPress theme. I did this simply by replacing the mini_rdf.gif image in the /public_html/wp-content/plugins/zitgist-browser-linker/imgs/ directory.
Then, also in keeping with my local theme, I made the text in the popup a bit smaller. I did this simply by adding a font-size: 80%; property to the style.css stylesheet in the /public_html/wp-content/plugins/zitgist-browser-linker/css/ directory.
And, that was it! Simple and sweet.
It is also important to realize that this is just a first-release prototype. Some initial bugs have been discovered and worked out, sometimes the server site is down, and longer-term potentialities are only now beginning to emerge. But, this is still professional software with much thought behind it and much potential in front of it. If it breaks, so what? It is free and it is fun.
To all of you out there new to RDF and structured, linked data, I say: Play and enjoy!
zLinks is only beginning to touch the most visible part of the iceberg. It is pretty clear that the use and usefulness of links are only now being understood. Harking back to the original listing of three possible uses for a link it is clear that “actions” and the use of the link itself as a referrer and “mini-banner” on the Web are still not appreciated, let alone exploited.
It is interesting that AdaptiveBlue has also come out with a SmartLinks approach that differs somewhat from the Zitgist approach (items and linkages are constructed and then referred to from a central location), but their screenshot does affirm the untapped potential of links.
The W3C semantic Web community continues to grapple with resource/link terminology and nuances, the implications of which will be deferred to another day and another blog entry. However, suffice it to say that with a growing ‘Web of data’ and linked data, not to mention the original document vision and then one of commerce and services, the once lowly link is growing mighty indeed!
