Book Idea: XmlHttpRequest Enabled Web Pages, or, Pull Web Pages

Since I’ve been writing the book, I’ve been having lots of thoughts about other books to write. It’s that same effect where, for example, someone points out a new type of shoe that you’ve never seen before, and then all the sudden you notice that “everyone” is wearing that shoe.

Push and Pull HTTP

Anyhow, I was looking through Steve’s links, and noticed that he had several bookmarks for XmlHttpRequest,
the JavaScript object that allows you to make HTTP requests in web pages, allowing web pages to pull content instead of just have web servers push content down to them. This is how GMail does it’s magic (according to Charles Miller…I haven’t dissected the source myself).

That all got me thinking that a book focusing on the use of “pull web pages” would be pretty damned interesting. It’d probably be about the size of the XML-RPC book. There’re more ways to do “web page pulling” than with the XmlHttpRequest, and there are, indeed, a handful of “patterns” and plenty of “gotchas” involved with it.

About 2 years ago when we were confronted with “GUI’ing” up our web interface at work, most of us outright rebelled at the concept of web page pulling: I must have said numerous times that it was heretical, or an abomination, or something like that. But, after having used GMail for about a year, I like the idea now. GMail’s so damn cool and works so damn well, and I want web applications I make to work be like it. Google’s new suggest feature is another good example of “web page pulling” that I think is really cool.

I don’t really have any time to work on a book like that until Chip and I finish
the JAAS book. But, if someone’s interested, I’d be thrilled to brainstorm on it. We could probably whip up a rough proposal and outline/ToC in about a week’s worth of calendar time.

Blogs Strike Again

“What’s fascinating about the Jon Stewart takedown of ‘Crossfire’ is not just what he said, but how his message got distributed,” Jarvis wrote. “The really stupid thing is that CNN didn’t do this themselves: ‘Hey, we had a red-hot segment…you should watch; here, please, look at this free download because it will promote our (hosts) and our brand and our show and give us a little of that Stewart hip heat.’ That’s what CNN should have done. Instead, they’ll charge you to deliver a videotape (what’s that?) the next day.”

“Jon Stewart ‘Crossfire’ feud ignites Net frenzy”

Big Time Blogging

First,
an article from Business 2.0
:

In theory, at least, blogs are a marketer’s dream. That’s because — unlike burning through millions of dollars on TV or print advertising campaigns — they are a virtually cost-free way to communicate with customers. And not just any customers. These are self-selected hard-core fans of a particular trend, hobby, idea, or product. “Bloggers are an incredibly influential consumer segment,” says Technorati CEO David Sifry. “These people are huge networkers. They get the word out quickly on products they like — and don’t like.”…The chief blog marketing goal, then: Create a community of knowledgeable insiders. “Done right, consumers will do all the marketing for the company — forwarding the information they found to their friend and encouraging others to visit,” says Lydia Snape, Internet services director for New York agency Renegade Marketing.

And then this note about Schwartz’s weblog:

Sun COO Jonathan’s Schwartz’s take no-prisoners style of blogging has apparently pushed Hewlett-Packard execs past their boiling points. According to a report on ITWorld.com, HP has written a letter to Sun demanding that Schwartz remove HP from the target list of his sharp-tounged Web log commentaries.

Fear in Software Development

In software development, fear is what drives much of the smoke-screening and mirroring that occurs. These tactics have lots of names like “CYA” (Cover Your Ass), or the generic “politics,” and they all negatively effect the development of software. Software development is primarily about figuring out what the hell it is the customer wants implemented, and then how to implement it, both code- and schedule-wise. Put broadly, software development is figuring out unknowns or, as they say, solving problems. It’s hard to solve problems when the set of positive data and other inputs you have is filtered and limited.

If it’s true that the negative stuff above is caused primarily by fear then, it’d be handy to know what those fears are so you can address them. Here’s a list from

Planning Extreme Programming
:

Customers are afraid that:

  • They won’t get what they asked for.
  • They’ll ask for the wrong thing.
  • They’ll pay too much for too little.
  • They must surrender control of their career to techies who don’t care.
  • They won’t ever see a meaningful plan.
  • The plans they do see will be fairy tales.
  • They won’t know what’s going on.
  • They’ll be held to their first decisions and won’t be able to react to changes in the business.
  • No one will tell them the truth.

Developers are afraid that:

  • They will be told to do more than they know how to do.
  • They will be told to do things that don’t make sense.
  • They are too stupid.
  • They are falling behind technically.
  • They’ll be given responsibility without authority.
  • They won’t be given clear definitions of what needs to be done.
  • They’ll have to sacrifice quality for deadlines.
  • They’ll have to solve hard problems without help.
  • They won’t have enough time to succeed.

(Thanks to Joe’s Wiki for the original transcription of the above.)

In addition to the human tactics (talking to people) of overcoming this fear, using tools that encourage transparency helps too. There’s a huge chicken-and-egg problem, of course, with those tools: if people are afraid, they have to get over their fear to start using the tools that help reduce their fear. My opinion is my usual “whatever <shrug> better to try than to sulk.”

Also, when check out this short post on transparency in the business end of things.

Transparency, XPlanner, Making Dates

Once I asked an IBM employee that worked on the Eclipse project (a dev) how they could turn around builds so quickly and he said roughly paraphrasing “my current boss, most likely my next boss, and all my friends know what I said I’d do because it’s right there in the specs on the web. If I miss my dates, everyone knows.” As much as some might argue that (as Tom points out in his article) “familiarity breeds contempt,” I believe that transparency breeds accountability, understanding, and ultimately more buy-in to what we are trying to accomplish together.

“Transparency, snowmobiling, and other potentially dangerous, but rewarding endeavors?”

This reminds me of something interesting Alistair Cockburn said in his ITConversations interview:

[P]eople find in the Agile projects there’s no place to hide. And you get the weak developers, the pompous developers, you get people who just fill space. They get pushed out really fast, because it’s “show up and contribute or get out of the way” and some people have said, “There’s just no place for the weak people to hide.”

We’ve slowly been adopting XPlanner at work — we’re just using it to track our backlog now. Essentially, it’s a scaled down MS-Project on the web, except it’s XP-centric so there’s stories, iterations, and all sorts of funky XP-talk instead of the usual project-management jargon.

Anyhow, what makes XPlanner “transparent” is that everyone can see what everyone else is working on, their estimates, and their progress. People tend to cringe at this at first. Indeed, I’ve heard some stories of XPlanner going horribly wrong: if you show this stuff to people who don’t understand software development (it’s not just 8 hours solid everyday of typing away at the keyboard…shocking, I know. I thought it’s be just like factory work too!) they start using it to track the coders in a bad way, and then people start putting in bogus data, and it’s useless as anything but a check list.

Back to the point: one of my more interesting philosophy professors had a pet theory that shame was an incredibly effective tool in keeping society together. In fact, he said, it’s not used as much as it could be. People can be motivated to Do The Right Thing to avoid feeling ashamed at having done something wrong, or, feeling ashamed at having failed to do something. Of course, it works ex post facto as well: if you feel ashamed at something you’ve done, you’re likely to fix it.

The connection between transparency and shame is put well in a phrase much favored now-a-days:
“sunlight is the best disinfectant.”
Which is to say, brining it back to software, with everything schedule-wise out in the open, as XPlanner provides, people are (probably) going to do higher quality work than if everything was locked up in an MS-Project file or spreadsheet somewhere.

Don’t confuse using XPlanner with the broader point here. That point being: in 90% of the cases in software development, the more knowledge that’s “public” in your team, the better off the resulting software and process will be.

Making Hay from the IM Stupid Network

As you may remember, AOL and Yahoo! closed
down their Enterprise IM a few weeks ago
. It

looks like they got a good deal from
Microsoft
: instead of having to create and support their own infrastructure, MS will
just allow their clients to talk to their IM Server, and:

Microsoft will pay AOL and Yahoo a royalty for connecting
to LCS. The companies declined to elaborate on whether these payments
will be based on the number of LCS users connecting to AOL and Yahoo.

You have to think this setup made it easier for the biz people to
tank their own IM servers. Now, instead of taking on the risk of
creating and marketing their own IM servers, they can concentrate on
their “public” IM clients. The more people that use the public
clients, the more money they make from the MS IM Server. Seems like a
good deal to me.

For Microsoft, it helps get over the tipping-point hurdle of IM in
the Enterprise: there’s a certain number of people you need using
compatible IM before it’s useful (and worth paying for). That
is, if you have 1000 workers, and 1/3 of them use AOL, a 1/3 of them
use MSN, and a 1/3 of them use Yahoo!, IM’s not really that useful. No
one’s going to want to lay out the cash for an IM server that only works
for a 1/3 of your people.

But, if all of those clients (or a majority of them) can
interoperate, you can get actual value from paying for the
server. And, of course, there’s all sorts of other features you can
build into applications (like Office), e.g.,:

if Outlook’s calendar shows a person in a meeting, it can
route voice calls to that person’s cell phone. Or, if someone sends an
IM to a user, the software can then prompt a Net phone call and record
a voice message.

Again, if only a 1/3 of your people can benefit from this, when a
sales rep is trying to sell you on Office 2050 with this feature,
you’ll say “so what, Office 2000 works fine, no need to upgrade.”
But, if everyone can benefit, then it’s a much better selling
point.

Anyhow, point being: it seems like a shrewd move, for all parties
involved, to allow their IM’s to interoperate. I guess it’s one of
those “everyone can benefit from the stupid network” things.

Wipro CEO Interview, Innovation, CMM Level 5, Input Bias, Workspace

  • eWeek Interview with Wipro CEO – interviews with offshore execs are always interesting. They’re becoming less defensive in response to the usual “you’re taking away jobs” question/claim, instead suggesting that it’s not their fault, but the fault of the country that’s leaking jobs, for example,

    To every one of the millions of unemployed people in the U.S., that job created in India looks like that was “my job.” So, there is an issue of a job lost here is a job created there. As a result, there has been an enormous magnification of how much impact outsourcing has had. The second is, as a result of the type of economic recovery being made in [the US], companies are not willing to open up the gates to hiring. Now you take that circumstance that in a political year it becomes a platform issue. It doesn’t matter whether it’s right or wrong, it’s a political platform. You will see this issue being politicized because of the perception of “that’s my job sitting over there.”

  • “The Offshore Proposition” – in a related column, Lundquist brings up the point that all this IT stuff might be too expensive (thus, prone to lower costs by offshoring) by design:

    Somewhere along the line, service stopped being, well, a service and started being a profit center. Once the bean counters took hold of the equation, they started to look for ways to charge users more for service while having their company pay less for it. The next thing you know, you’re talking to someone in Manila about your cranky computer. Designing systems that don’t need service because they work as expected would be the best service offering vendors could develop.

    And, as alluded to in the above interview Vivek Paul, if offshorers can create such services, they’ll start their own product lines,

    Once the offshore companies realize they have the business development, design, manufacturing and service skills, they’ll begin asking themselves just what their U.S. partners are bringing to the table. This will bring about the emergence of new competitors among companies that were once partners, which, I predict, will be a hallmark of the technology market over the next year.

  • Bursting the CMM Hype:

    [W]hat matters is what’s behind the impressive-looking number. Is there a verifiable commitment to quality, process and training? Can companies demonstrate improvements they’ve made over time in customer delivery times, developer productivity and defect density? Will the project managers that went through the assessment be assigned to your project? If the answer to any of these questions is no, then a CMM Level 5 isn’t worth much.

  • Bias Beware – an interview discussing “input bias,” how the selection of information you’re given can lead to incorrect results. The ideas mix well with the bike shed by committee problem, for example,

    People’s judgments often depend on how easy it is to evaluate something. If it’s easy to measure the outcome, we may not rely on input measures. If someone is performing in the Olympics, their time is a clear measure of how they did. But in other cases, such as judging how innovative a pharmaceutical company is, it’s harder to reach a decision. So you might rely on whatever objective measures are available, such as how many patents the company has. But in reality, that’s not a good measure because more patents are filed for small modifications of existing compounds, and there aren’t really that many big blockbuster drugs involved.

  • Workspace Survey – most people give their workspace a C.

Links: Open Source Research, Dell Systems Management, Google Stickyness

  • Research Shows Effectiveness of Open Source Model of Developmentthe homepage for the research group has lots of interesting looking papers. The othe research groups at the ISR look interesting as well.
  • Dell Pushes Management Standards:

    [Customers] have to choose to invest in multiple proprietary vertically integrated management stacks from each of the vendors for their hardware in parallel to the management they’re using to manage their operating system and application deployment. That increases their costs of managing the environment. If they don’t do that, then ultimately they are locked into one or, if they’re lucky, two vendors, and have to live with the decisions that vendor makes, which won’t always be optimized to the business problem that they’re trying to solve inside their industry.

    We [in the high-tech industry] have tried to out-innovate each other, but ultimately that has driven up the cost of ownership for customers.

  • Google Stickiness, Command Line Interface to the Web:

    Google has decided that its customers should gather information through inputs of text search terms by using more or less the same simple interface to search for news, things to buy, or any other topic. That’s a small but important distinction. Google assumes that customers are smart enough to learn to search with words rather than with the graphical and pull-down menus used by most of its competitors. That’s an understandable bet. Google has gone from upstart to Internet star with a business plan based on that assumption.

    This is an interesting point: Google really is a command line interface (CLI) to the web. Well, of course, a sort of modern day CLI that’s done through a web browser instead of a shell. Nonetheless, it’s wide use and popularity gives hope that the optimist have been right about everyone being able to use CLI’s. They’re so much quicker (at least, for simple tasks) once you figure them out.

    Also of interest, and alluded to, is the extremely loose coupled architecture of Google’s services. To “integrate” with feeds and UPS tracking they didn’t have to exchange code with either of those companies, Google just used the most brilliant and simple RPC ever: the URL…and, implicitly, HTTP.

    There’s also this interesting marketing note:

    At the same time, being included among Google’s services could become a key selling point for companies such as FedEx. It’s not that much of a hassle to go directly to the DHL Web site to track packages shipped on that FedEx competitor. But when anyone can check package status directly from the Google toolbar, that constitutes a nice little edge for FedEx or UPS. In the brutally competitive package-shipping business, that small advantage can have a big impact.

Links: The Future, Prototyping, Combat Metrics, and Chinese Standards

  • Marc Benioff, of SalesForce.com talks about The Future – another of those “tell us about the future” interviews…quite long, though.
  • Prototyping Toys:

    Two-and-a-half-years ago, Little Tikes could not show off a toy until six months before it hit the shelves. The company now demonstrates what the product is supposed to look like as much as a year ahead of time using stereo lithography. The difference, according to Little Tikes president Rory Leyden, is that only 60% of new products used to meet the company’s profit projections. Now, two-and-a-half years later, the success rate is 96%.

    Also of interest is that their documentation process is composed of MS Office files (Word, Excel, etc.) and SharePoint.

  • When Using Metrics is Wrong:

    “The issue of quantifying success in counterinsurgency operations is a fool’s errand,” said one officer based in Baghdad. “It is great for business management, but not for the conduct of war. It is something that is questionable in conventional warfare and downright dangerous in unconventional warfare, simply because it will force you into taking actions based on that which is to be measured and not on what needs to be done.”

    (Link from Ascription is an anathema to any enthusiasm.)

  • More In Depth Discusion on China Making it’s own “Standards” – as I’ve linked to other posts, China is starting to make up it’s own standards. Also found at Ascription is an anathema to any enthusiasm.

IRS Lost it’s Wallet in El Segundo, Finger Pointing Theatre

There are several articles about the IRS’s IT revamping, done by outside contractors lead (?) by CSC, getting out of hand 2 years into a 5 year schedule. They’re not very in depth, e.g.,:

“Both the IRS and CSC get blamed for failing to establish an environment of trust, confidence and teamwork between themselves. In fact, the report states, the opposite is true, leading to a tense and inefficient working environment and regular finger pointing. The report also criticizes the IRS and CSC for laying down an unrealistic program schedule.”

I can’t find The Report either at the IRS Oversight Board Website, though, there are some other interesting looking things there.

CNET has more (from the NYTimes):

The IRS says it can still process returns and send out refunds on time, but its dependence on the 1960s-era Assembler and Cobol computer languages makes it difficult to investigate and resolve taxpayers’ problems. Finding a record using the existing system can take a week; the new system is supposed to do the job in seconds.

. . .

Paul M. Cofoni, president of the Computer Sciences unit running the project…, said “in the early part of the program, we did a poor job of defining” what needed to be done. But that was in large measure because the IRS had no records of many changes to its old system, he said, and was reluctant to approve specifications for the new system until it could be sure that it would be able to find and display all the old information.

. . .

Five years into the project, some aspects are as much as 27 months behind schedule.

It looks like they’ve encountered Glass’s top two reasons for software project failure: missing requirements (and the inability to adapt to “finding” them re: schedule and over-all project re-jiggering), and, bad estimates (and the lack of revising the almost usefullness initial estimates throughout the projects).

Re: Software Facts and Fallacies


As always, Arley has fun (no sarcasm there), lengthy comments on our favorite topics. This time, on today’s post about
Robert Glass’s Facts and Fallacies of Software Engineering
:

The book is okay, but it’s the same as all the other software engineering books: here’s what you shouldn’t do with all the examples of what went wrong, and here’s my theory of what you should do with absolutely no empirical basis whatsoever.

In his defense, many of the comments (well, that I’ve read so far) draw on successful aspects of projects, not just “do the opposite of this bad thing.” But, you’re right, it would be nice to read more accounts of success rather than accounts of what didn’t work out. And, yes, a book like that would be rad. I think Cockburn’s writing (fadish as some of it is…well, fad-setting) is of this bent as well. Also, in the book
there are some references to a NASA book of software management that looked interesting in this success-only vein.

The argument about the top performers and office space / working environment is completely fallacious, and you as a philosopher should easily recognize it as such. The author makes an implicit and unjustified assumption that better office space will result in better programmers. You’ve got to look at the cause, not just the effect. Perhaps the better programmers have better working conditions because they’re better programmers?

And, sure, sure, the “more space makes more productive workers” is not a rock solid (hell, even of wet noodle strength) argument: I immediately had the same thoughts as you. What you’d want to see is a group of people who were performing badly, move them to better workspaces, and then see them improve…and then repeat it until even Bacon would say, “enough already! Let’s have super!”

Personally, I don’t think office space has so much to do with benefits to productivity (directly), but to moral. Sure, we can say that’s an egotistical idea–it is!– but stroking egos (when they deserve it) is a huge part of keeping up moral.

To pick up on the success-only basis for broad, unprovable statements, the most successful projects I’ve worked were staffed by people with extremely high moral who were always very happy. If that moral and happiness ever dipped, the quality of work dipped too; once moral and happiness went back up, so did quality and productivity.

The problem, of course, is often that the situation is more complex: if ther’e someone who’ll take your place (either on- or off-shore) when you get fired for complaing about moral and happiness, and you can’t afford to loose your job, then you’re shit out of luck, you better shut-up and code. Whether that works in the long term is often of little worry to orginizations who work under similar fears on a quarterly basis.

(And, as one of my co-workers, Brandon, can probably appreciate, I just thought it was a good excuse to use a picture I’ve had laying around for some time ;>)

Coffee in Bed, or, Thanksgiving Lazy-Links

  • US State Governments avoid offshoring, e.g., “New Jersey now spends $340,000 per month on supporting its welfare benefits applications from a call center in Camden, which represents a 28% premium over the $266,200 monthly charge it was paying for the service when it was delivered from Mumbai. ” There are interesting responses to The Register’s article as well. Also, the CNET story on the same topic.
  • iPod Users Strike Back! “Irked at what seemed to be the early obsolescence of the music player, the brothers trekked around New York City stenciling the words “iPod’s unreplaceable battery lasts only 18 months” on all the iPod posters they could find.”
  • Chinese Myth And Reality:

    The way to beat China is to stop being scared of pressing our advantages, and to press them. Our advantages are speed, capital and liberty. We have tossed those away to fight the “war on terror” and in doing this, we have given away the real game.

  • On not using RSS/Aggregators: “I think the most important thing about this technique is that I am reading a wider variety of sites now than I was with RSS. With RSS, at a certain point, I was basically slaving away at reading all of the sites I was already subscribed to. Without it, I try out new names on blogrolls more often. “

Outsourcing Business

Evalueserve provides services for patent writing, evaluating and assessing the commercial potential for law firms and entrepreneurs. Its market research services are aimed at top-rung financial services companies, to which it provides analysis of investment opportunities and business plans. Another major product is multilingual services–Evalueserve trains and qualifies employees to communicate in Chinese, Spanish, German, Japanese and Italian, among other languages. That skill set has opened market opportunities in Europe and elsewhere, especially with global corporations.

Experts say these new trends are significant, and they will continue to grow over time. “Activities considered for ‘offshoring’ have moved up in value and begun to touch core functions, such as highly analytical processes,” says Stefan Spohr, a principal in the financial institutions group of A.T. Kearney, a global management consulting firm in Chicago. “More complex customer services are substituting simple data processing and call center activities.” Spohr adds that the higher-end functions being performed offshore these days include information research, financial portfolio analysis, customer data mining, statutory reporting and inbound insurance sales, among others.

. . .

Cutting costs is not the only reason why outsourcing such tasks makes sense for its clients; it’s also about higher quality of work, says Aggarwal. “Among the more unusual emerging developments is that business process offshoring is not merely a way to reduce cost by migrating core functions,” adds Spohr of A.T. Kearney. “It is also a strategic initiative to take advantage of technological advances and the human capital available offshore to fundamentally restructure an organization’s operating model.”

Evalueserve’s model works on a mixed system where anywhere between 50 percent and 80 percent of the work is handled out of an Indian facility, while the rest is done at the client’s location. For example, a patent filing assignment from a U.S. corporation may involve the Indian staff writing the patent in English or say, Japanese, and evaluating its commercial potential. But the client or its law firm would do the actual filing in the United States.

Us coders carp about outsourcing/offshoring all the time. Among other things, we always think it’s witty to suggest outsourcing CEO’s, managers, and other business positions. Well, there ya go. Yay capatalism!

Outsourcing Business

Fowler on “Enterprise Architecture”

As it turns out, I can get pretty cynical about enterprise architecture. This cynicism comes from what seems to be the common life-cycle of enterprise architecture initiatives. Usually they begin in a blaze of glory and attention as the IT group launches a major initiative that will be bring synergy, reuse, and all the other benefits that can come by breaking down the stovepipes of application islands (and other suitable analogies). Two or three years later, not much has been done and the enterprise architecture group isn’t getting their phone calls returned. A year or two after that and the initiative quietly dies, but soon enough another one starts and the boom and bust cycle begins again.

So why does this cycle happen with such regularity? I think that most people involved in these initiatives would say the reason they fail is primarily due to politics – but what they often miss is that those political forces are inevitable. To succeed in these things means first recognizing the strength of those political forces.

. . .

[Duct Tape Architecture?] A good way to think about this is that these initiatives should less about building an overarching plan for applications, and more about coming up with techniques to integrate applications in whatever way they are put together. (After all ApplicationBoundaries are primarily social constructs and they aren’t likely to conform to anyone’s forward plans.) This integration architecture should work with the minimum impact to application teams, so that teams can provide small pieces of functionality as the business value justifies it. I think you also need to focus on approaches that minimize coupling between applications, even if such approaches are less efficient than a more tightly coupled approach might be.

Be sure to check out the “Patterns and Best Practices for Enterprise Integration” link.

Fowler on “Enterprise Architecture”

Microsoft System Management

Microsoft does not plan to take on systems management heavyweights such as IBM, Hewlett-Packard and Computers Associates. But management software has become a key element of the company’s plan to convince corporate customers to base their most important systems on Windows, Hamilton said.

. . .

Microsoft’s two main management products–SMS and MOM–differ in capabilities. SMS is geared toward letting large companies distribute software updates and patches automatically to PCs over corporate networks. MOM, meanwhile, is for monitoring network events to head off problems, such as an overloaded server or dropped network connection.

. . .

With DSI, Microsoft is seeking to automate many data-center operational jobs and reduce the labor involved. The idea is that management software can be clever enough to know when a given application will have a problem and take actions to avoid it. For example, the systems management software could fire up an extra Web server when the existing machines are being overloaded because of a spike in traffic.

. . .

Central to DSI is an Extensible Markup Language (XML)-based data format, or schema, called Systems Definition Model (SDM). Microsoft calls a SDM a blueprint, or description, for how software and hardware components can be controlled.

Microsoft System Management

META keywords for Whitehouse.gov

“social security, medicare, energy, tax relief, Education, policies, White House history, White House news, news, United States of America, 43rd President, George W., W., George W, President George W. Bush, President Bush, White House, government”

Usually, the old <META name=”keywords”/> HTML tag is used to describe what the page you’re looking at is about. Search engines (supposedly) use them for better indexing too. This META block may be old, but it’s interesting to see what keywords the White House associates itself with. Most of the Clinton Whitehouse.gov pages don’t seem to have the META tag.