Posted:April 13, 2010

You’ve Got to be Crazy to Look to an Ad-based Revenue Model

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!

Posted by AI3's author, Mike Bergman Posted on April 13, 2010 at 10:40 pm in Uncategorized | Comments (1)
The URI link reference to this post is:
The URI to trackback this post is:
Posted:February 2, 2010

Inkscape Logo

The Inkscape Process Can Also Aid Image Interchanges with Powerpoint

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.

Why 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:

  • I wanted re-use of image components via re-sizing and re-coloring, etc., and vector graphics are far superior to raster images for this purpose
  • I wanted a stable, free, usable editor and Inkscape was beginning to mature nicely (the current version 0.47 is even nicer and more stable)
  • Its SVG (scalable vector graphics) format was a standard adopted by the W3C after initial development by Adobe
  • SVG is an easily read and editable XML format
  • There was a growing source of online documentation
  • There was a growing repository of SVG graphics examples, including the broadscale use within Wikipedia (a good way to find stuff from this site is with the search “keywords site: filetype:svg” on your favorite search engine, after substituting your specific keywords).

How to Collaborate with Inkscape

Once you have a working image in Inkscape, make sure all collaborators have a copy of the software. Then:

  1. Isolate the picture (sometimes there are multiple images in a single file) by deleting all extraneous image stuff in the file
  2. From the toolbar, click on the Zoom to fit drawing in window icon [Zoom to fit drawing in window]; this will resize and put your target image in the full display window
  3. Under File -> Document Properties … check Show page border and Show border shadow, then Fit page to selection. This helps size the image properly in the exported file for sharing or collaboration
  4. Save the file as an *.svg option, and name the file with a date/time stamp and author extension (useful for tracking multiple author edits over time)
  5. If in multiple author mode, make sure who has current “ownership” of the image is clear.

How to Share with Powerpoint

Of 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:

PNG representation of an SVG

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):

PNG representation of an SVG

We can save it as an EMF (figure_textpath.emf) to a Powerpoint, with the option of converting text to paths:

Text-to-path EMF

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:

EMF with no text to path

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.

Other Alternatives

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.

Posted by AI3's author, Mike Bergman Posted on February 2, 2010 at 11:26 am in Uncategorized | Comments (0)
The URI link reference to this post is:
The URI to trackback this post is:
Posted:October 1, 2009

Friday Brown Bag Lunch

Serving Up Occasional Re-treads and Leftovers

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!

Friday Brown Bag Lunch 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. ;)

Posted by AI3's author, Mike Bergman Posted on October 1, 2009 at 4:24 pm in Uncategorized | Comments (0)
The URI link reference to this post is:
The URI to trackback this post is:
Posted:February 29, 2008

Resurrecting Old Posts Brings a Smile, and Some Shudders

(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.

Why the Upgrade?

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.

The Benefits (sort of)

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! :)

The Problems

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. . . .

Advanced TinyMCE

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.

Posted by AI3's author, Mike Bergman Posted on February 29, 2008 at 12:44 am in Uncategorized | Comments (0)
The URI link reference to this post is:
The URI to trackback this post is:
Posted:September 29, 2007

zLinks Kicks Out an Old Favorite

zLinks from ZitgistThe 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:

  • different link colors than the original Web 1.0 link blue,
  • sometimes no underlining,
  • sometimes dotted underlines,
  • even boxes, and (horrors!)
  • even upper and lower borders!

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.

zLinks Raises the Link to the ‘Power of Z’

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.

A Diet is the Only Cure for Iconitis

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:

Example Link Icons

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.

The Scope and Longer-term Paradigm Remains Unclear

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:

zLinks Popoup

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:

zLinks Popoup

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!

Posted by AI3's author, Mike Bergman Posted on September 29, 2007 at 2:13 pm in Uncategorized | Comments (0)
The URI link reference to this post is:
The URI to trackback this post is: