 So the concept is the following. I think that in the end we have been used to think about in the end for about four months. And if we compare the four months then we start getting into complex texts in our discussion which are always based also on opinions. So you can say that doing that kind of feature in that way or in the other one is better or worse and that is technical. I think that ODF has to be considered as a standard because what is absolutely true is that ODF is a real standard while Office of an XML is not a standard at all. If we consider that under the standard point of view and we completely ignore the fact that it can be or not a good format. So we open source is shared knowledge and we have a couple of two standards. So ODF is a logo, Office of an XML is not a logo. Someone decided with bananas and I think the appropriate logo is Office of an XML done with bananas. You should consider additionally that the only thing that I do not eat is bananas. So for me it's disgusting in terms of personal feeling. So as you know that I'm not technical so I try to give you an overview as a let's say common user of the four months. So let's take the ODF in general. So ODF, there's ODT but it's wrong. So it should be ODF daily brown face. The inner complexity that you find in the files is very low or no hidden complexity at all. When LibreOffice writes Office of an XML the approach is the same because of course the way that the software saves the file is more or less the same. So I can say that an Office of an XML file written by LibreOffice is cleaner actually of the same file written by Microsoft Office. And files are human readable and this has a lot to do with security. An Office of an XML file written by Microsoft Office has the highest option of inner complexity and this is done on purpose of course. Because it's more difficult to hide the complexity that not with the complexity inside the file. Same approach when writing ODT. So an ODT written by Microsoft Office apart from the issues in terms of application of the standard is more complex than the same file written by LibreOffice and files are not human readable. The consequence of these two lines is that if you look at the database of vulnerabilities of the National Institute of Standard Technologies which is an organization of the American of the US government. So I would say that it's neutral to close to proprietary software. And if you look at the last three years vulnerability is referred to ODF R11 vulnerability is referred to Office of an XML R194. And this has to do with the fact that not being human readable the file can hide whatever you want. Why not find that it's human readable if you tweak that file it's a lot easier to spot where the file has been tweaked. So if you insert for instance a same link in a LibreOffice file in a good ODF file. The same link is a lot easier to spot than the file than the Office of an XML file. But at the end of the speech you will understand why. So this is the you know computer system a neat and complex thing to normal use. But let's say that why are you getting slower in the case of Office of an XML is at the highest level. And by the way people tells me if this is Photoshop. No this is a Microsoft manual that means how to lock in your clients. If you want it I can send you the PDF it's easy to find on the web. And it's a manual it's 2004 but no one since 2004 I said oh we brought a stupid thing in 2004 and now we are not locking clients anymore. The difference is in the process. When you make a standard there is a process. So this one look at the timing 2002 and then 2006 it was approved. But the 720 pages of the standard were reviewed in 1239 days. So it means that more or less one page every two days. In this one you have three years. So one year less the standard is 7200 pages that have been reviewed in 838 days. So it means that 100 pages of review per day more or less. And this is not piece and work by Tolstoy is a description of a standard. So you can read it with the same speed of a novel. So basically that means and this is evidence that means that the standard has not been reviewed. And so it's not a first let's say first thing that is not a standard. Second one which standard do we use for the FWCOR, XLS, FO, SVG, MATAMEL, X-Link, SMAEL, X-Force. So where there is a standard ODF reuses the standard. Of course this has to do with the lower number of pages of the description because of course you don't re-describe MATAMEL. You just say go to the MATAMEL standard and read the manual. Here the only standard that has been reused is WCOR. Otherwise it's not XML because WCOR are the specification for the XML. So if you want to write an XML document you must reuse the WCOR. So they reused the only standard that they were forced to reuse. They have not reused another standard, secondly that is not a standard. From right on the far stack this is a DOC file. The PC, MS Windows, MS Word and DOC. And this is the pseudo standard file style. PC, MS Windows, MS Word and then you have the same. So there's no, you know, to write an interface, to write a Windows SDK, the XML specification. The only thing that is close to standard are the XML specification. All the rest of the stack is programmed with that. ODF, ISO standard, ISO W3C standard, ISO W3C standards. Of course this doesn't mean that ODF is perfect but at least it's based on standards. So it's reusable, easy and so on and so forth. Let's look inside ODT. You know that both ODF files and Office of an XML file are ZIP files that change the extension. So if you change the extension to ZIP and you expand it, you get a folder. And inside the folder there are documents. So this is the folder of an ODT file. Try to visually remember it. You know, the key element is content XML. That is the key element. Inside the documents, first, two levels, quite two levels. Everything is inside the world. So let's say that one of the concepts when you do a standard is that a standard should be, especially if a standard covers different formats. As in the case of the Office of an XML, it should be consistent for the different formats. So the structure of the files should be exactly the same. So this, we have two levels here. Okay, let's say that two levels are a very nice thought. So two levels are better than one. In Italy there is, we say that there is two levels are better than one. It's an ice cream and advertising. But this is not, it's good for ice cream but not for standards. ODS. You know, some people were so courageous to tell me that ODF is boring because it's always the same. Because an ODS file is exactly the same as an ODT file. And I said, this is the advantage. Because if something changes it, you immediately understand that it's changed. Three levels. So not, three levels are better than one. And because here you have a worksheet. What is, in the case of DocEx, there was Word. Here you have Excel but not Excel the entire, but just Excel. So another way of not being a uniform employer. And then you have a worksheet. And then you have a sheet one XMM where you find all the contents. Go to ODP. Again, boring. It's the same. You always find media and pictures. Because in the presentation you have pictures and multimedia. And of course you need two folders to store pictures and multimedia. But again, the structure is the same. Content, meta, mankind, setting, styles. Easy. And also these files are consistent. So the structure of these files is almost the same. Three levels. But here you have another difference. Because apart from the fact that it's the first level is PPD and OK. Then you have slides. And here you don't have a single file. But you have a fine per slide. So again, 13th that is not a standard. Because it's not uniform. It's not coherent in terms of approach. So if it was two levels for the three and two levels and one document for contents for each one. You could say it's not optimal as a standard. But at least it's coherent in its different way of being applied. This is not, I mean, not only they have a different approach for each format. But then in this case, for instance, you are, let's say that the presentation wraps completely. And you have to unzip the file to recover all the contents. No way of knowing. Let's say, I mean, this was a nine slide presentation for a test, which is just a very short one. But let's say that you have a 100 file slide presentation. So you have 105 here. And you don't have any correspondence with the media and slides folders. So there are codes, but not names. So if you have 100 images, you have to dig into 100 codes and not 100 names of images. In the case of the F, you have image one, image two, image three, image four, image five. So slide one usually as image one and so on and so forth. People that have seen my presentation know this very well. So this is the, luckily we don't have women here, but not because I hate women, but because you cannot discuss about colors with women. You know, men have 16 colors in their mind and we are fine. We are all 16 colors, this you have for everyone. And for a computer is FF00000. And we all agree on the concept. Okay, again, ODF is boring because writer says format, color, FF00000. And count doesn't say, impress doesn't say. Word, W, okay, word W. You have a little bit of color. W again, Y value is the first one. Except color, RGB. It's not an RGB color, by the way. And then you have 4F and 4 zeros. PowerPoint, A, YA. I tell you, because the program when written by the original software house and it was built by Microsoft, it was starting with A. So RGB, color, value, FF00000. Okay, so you have 3 programs, 3 ways. No one of them respect the XML syntax. The only way of describing colors. I mean I could write, there is a red here. It would have the same meaning. This is not standard. These are not standard. F, O, semicolon, color is a standard XML notation. None of this is a standard XML notation. No way, if you look at XML, you don't find this only color. You find only color, but not with W before. Because the color is a format. And then you must have an F, O, semicolon, color. Again, of course all this stuff is not known by users. And users do not mind about it. But if you put that in context, then users start to mind about it. Especially users that are in charge of choosing the standards. Then there is the nice meaning handling of calendars. So the data never existed. February 29, 1900. People have told me, okay, but why do you care about the data that is so old? And then I show the chart of the Italo's life milestones that are just after a few slides. And then they understand why. And LibreOffice is not it. I tested that, I know you did a lot about this. Yes. And you probably did it. I tried that with Excel 2016. And it doesn't do better. Not at all. That works. It works, but not if you take it and you take all the days that screws up Excel again. It's an option. You can try it out. Yeah, but still it was the one to work with this model not. Yeah, sure. But when I tested it was with Excel 2013 and it was a nightmare. That was always a nightmare. Today it works if you are in control. But for a normal user it would be a nightmare anyway. Because it changes and screws up things. And by the way, if you save this as strict it will delete this one. But it would not delete the day. So what is on March 1st, 1900, will become a yes, maybe shift all the days one day. Because I said this because my grandmother, I don't know if it's just by chance, was born on February 28, 1900. So I did all my grandmother's life as a single spreadsheet. And then I saved the script. And my grandmother result, the birthday result in February 27. Because they shifted all the days. And Microsoft does not yet reply. Because every six months I send again the spreadsheet to Microsoft asking. Did you, I mean, my grandmother is still asking me from, you know, she was, she died a few years ago. But she's still asking me to fix Excel days for her life. Because she's not happy of being born one day before. Now it works in some cases, not in all cases. So in the past, the network days was given as weekends, Saturday and Sunday. Which doesn't make me very happy. Because it's due and for them the Sunday, let's say that the Sunday is Saturday. And for Muslim, the Sunday is Friday. Which is perfectly okay. Because it's just a convention. The fact that you pray on Sunday or Saturday or Friday, it's up to you. And it's not up to the calendar. But Excel made it up to the calendar. Now they fix it partially on 2016. But not completely. But these have not been fixed. So language code have not been fixed. They use language code that are not standard. They don't use the ISO standard. They have issues with graphics because they refuse to implement SVG. So they have Windows and WMF as a vector format. But that has been, in addition, they have a 3C standard. And they have their format. Which is even non-compatible with Man-M-M. And again, Man-M-M is part of standards. They create no standard codes. These are the SVG codes. And these are the Office of an XML codes. This is very nice for the Italians because it's a cop-a-couser. That means that this library has a temporary work contract. So sometimes in the future it will change. And let's do some fight comparison. So this is a very stupid document. Two pages is not an ipsum. Don't try to interpret that. It was done on purpose to avoid any kind of potential influence of the dictionaries. So Lora ipsum is treated as foreign language by everyone. I think we are at least the standard Lora ipsum. I think we are in some world. At least a bulletin list and a table with the last line in the table is merged sets. These are the results. So any version of LibreOffice is 2022. I must say that once it was 2023, I do the comparison every three months. So in one case it was one line more. If that document is written by Office, as you can see it is more than twice the size in lines. Office 2010 transitional, it's 1040 lines. Same document, same contents. 2011 transitional, that was the Macintosh version. It's 12,854 which is a worldwide record. I don't know if it's good for the Olympics or an Olympic record as well. 2013 transitional, it was 1,590 lines. Same document, same XML. So no changes, pure saving. Save as create from scratch. I have a text document that I import every time. I apply all the balls and stuff. It takes two minutes. And I save it as dokex. 2016 transitional, 11,667 and MacOS 11,646. According to Microsoft declaration, the format of the two version is exactly the same. Okay, only 20 some lines of difference. But then a real stupid guy and I do the saving every three months, digging more. They have a seasonal way of saving the document. So 2013 transitional is 15,000. But in late 2018, it's 13,515 lines. The same document, same software, same machine. I use a virtual machine. So the virtual machine is a Windows 7 virtual machine. It's always the same. Same configuration, same Windows license, same everything. There is only on the machine. There are only stored Microsoft Office and Libre Office. So you cannot have any influence by other software. Office of Web 7, 2016 transition in 2017. It's 11,667. But in early 2018, it was only 960 lines. And they probably realized what they did. So in late 2018, it's again 11,288. Macintosh. 2017, 11,646. Late 2018, 854. I have all the documents on my artists. I'm happy to share them with everyone. This is the Alceto Cold Standard. A format where the XML, which is the core of the format, changes to that level. I mean, if it was 222 to 223, let's say even 250, you can say maybe some way of parsing the XML changes the number of lines. But going from 854 to 13,515 for the same document is just impossible. Let's pretend that Shakespeare writes in XML. So, sorry, start text, it would be just text. As I showed these to people that has no knowledge, I used start text and text. Just a question. Yes, yes. On the stricter, actually there is more point, which is not the latest, where the stricter and transition were identical character by character. So that means that one was not the other. Technical, I don't trust myself on this. But the problem is that no user is using stricter. For example, let's say that I have to dig a little bit on stricter and do a better comparison because with stricter things change a little bit, not completely. So, of course, start text would be only text and text would be backslash text. But I showed that to students of six years old. Not showing them the XML. So this is Shakespeare according to Libre Office. And this is Shakespeare according to Microsoft Office. So, two, it's, of course, then it tells you to keep together because then there is a single line with a space and then B and so on and so forth. So this is the way that a simple, I mean, to be or not to be, this is the question, I mean, it's, I think, one of the simplest sentences known in the world. And you go from one line of XML to, I think, 40 or 45 lines of XML. Of course, the reaction, technically, that is this is all correct under XML syntax. Yes. But why you do it so complicated? There must be some reason behind it. And I will give you the reason. But then what is really amusing, this is about Excel. So these are the, what I consider my life myself. So that was first, I brushed my nose, I got a degree, I got a first job, I got a PC, I got married, I was featured on BBC once in my life, I installed an office, I repair my nose, I launched LibreOffice and I got married again with the same life, not being divorced in the meantime, which I consider very important. So according to LibreOffice, these are the days which are human print books. And according to Excel, these are the days which I think really... Because to Excel users say, you should know at least your birthday in Excel data. So my birthday is 1948. And I'm proud of being born on the day 1948. But ask someone to make a data out of that. And this is Excel, this is an Excel file transitional. No, this is what 99% of Excel users exchange every day should be forbidden by all because at date, on the other hand you can tell to a government, I store your day in that way. It's just something that is unasshappable. I'm a stupid user but I would not trust a program that stores the date in that way. Because this is an algorithm, of course, and in a program that is based on algorithm, if you calculate the date with another algorithm, I don't know, I mean anything may happen. So deduction of a very stupid envelope, LibreOffice developers, and a close-off to the one that we have here, so to Armin, a bigger close because he's a genius. He's a geniusist and Microsoft has developed a bunch of assaults they don't know how to develop. But I think that the fines are stuck with something that is completely useless just to reduce the chances that software that is not aware of the algorithm because if you check the algorithm, based on the day of creation of the software, of course you can embed that algorithm into the software and the software checks the day and according to the day opens the file in a certain way. It's easy. Even someone that has a degree in literature like me can understand this simple equation. So the issue is there. They still have a turnover of 25 billion dollars based on Office of an Excellent Files. So they will never allow anyone to decently open and save those files, which I think is an issue for all of us, is an issue in terms of cost. And unfortunately, the cost is very difficult to calculate. But we had some luck. So the National Institute of Standard Technologies was approached by the US Capital Facilities Industry. These are the people that give to people that are building skyscrapers or large buildings in the states, large department stores, huge buildings which need a huge amount of money. So usually when you build such a building you never pay cash up from everything. You go to someone that gives you the money and then when the building is finished you give back the money. But of course these guys are not stupid. And in all the money that they were giving, they were asking of course for some evidence of the expenditure and one billion dollar was missing. So they asked the National Independent Party, they said please make a research where we are losing that billion dollar. Because a billion dollar is a billion. I mean, I need a billion dollar to leave each year. So I would lose that billion dollar would be worrying. And they made a 67 page document happy to share with you that is called Cost Analysis of Inadequate Interoperability. I think that this is clear in terms of so if you don't share interoperability this is analyzing what lack of interoperability is caused. Okay, that translation costs only two million dollars. Okay, negligible. Manual re-enter cost 462 and something dollars. Reworking design files, less than one million dollars, one million dollars, that is acceptable. Let's say that related, this is before the work start and this is after the work start. Of course I'm summarizing the document but I'm really willing to give you the document. You can read it. There is an entire paragraph explaining what there is in the table. So this is after the work has started. You have another 368,000 dollars and 27 million for manual re-entry cost. Manual re-entry means that the document was not readable so they had to re-enter the document. So half of this 500 million dollars per year are lost by one single industry, one single industry. Independently how big it is but one single industry loses 500 million dollars per year because of interoperability. So interoperability is a huge cost for companies. You can imagine what happens in other industry banking system where you have contracts. You cannot read the contract, you have to re-enter manually the entire contract. Insurance. You have all the policies. On all policies you lose the document, you have to re-enter manually the entire policy. Of course it's hidden cost, not visible because this is people that are working at their desk. But this is what interoperability is causing. It's causing a lot of lost internal money. And I think this is important. I'm happy to work with any of you to promote this kind of information. Of course this is kind of a marketing presentation. It's not technical. Technical people would be horrified by what I say but I don't mind because when I talk to people and when I talk to Italian politicians they usually at the end they don't say anything then when I exit the room they follow me and say but it's everything true. And I said I can't send you the documents so it's everything true. I have all the documents all I have to show you is based on facts, not on invention. And luckily we had this research by the National Research of Standard Technologies because we were missing any data on interoperability cost questions. Maybe it's too late because I speak too much. I don't know yet. So if you have a question I'm happy to answer the question even if we are late. I didn't say anything. You should be all with me. We might upload the test documents to talk about which go larger and larger about positions of various bungalows and provide TURL to the bungalows like some good life or something which is provided with free space. So you might keep any URL where somebody can download these documents because I'm serious. I would create a zip where I would have everything to get inside. Exactly. No. I prefer... Sorry. No, I prefer to send it... You know, I know my mind is of friends. No way. No way. They have to... Before I give that for free. Okay, then just send it. Because I spent dancing with that. It didn't come out in five minutes. Okay, I'm very curious about that. Because that is... I think, you know, unfortunately and I think this is part of the general issue Microsoft is extremely good at doing marketing and the fact that they are not doing marketing of Office of an XML is a sign because if they have something good we would be bombarded by marketing of that. They would do it every day. The fact that they have stopped talking about Office of an XML I think three, four years ago when the real issue came out shows that they know that it's totally crap but of course it becomes their position. So... They don't use it. The strict. They use the... They use the transitional. You don't find strict documents. But I have to be further and I will do that during the next quarter because I need to really... Because the problem with the strict is that in some cases they take the transitional and they tweak the transitional to make it strict. So in the process they use some pieces inside so you still have some transitional content in the strict document. Because I mean, I showed you a two-page document or a short Excel but just take a normal Excel of five sheets for 1000 line-inch which is rather normal in a company. So that means that probably that Excel would be 200 or 300,000 lines of XML and to convert that from transitional to strict you can do some bits in 300,000 lines. It's very easy because even if you have a very good algorithm Boy, the way that they write it they repeat the color of a line, every line of a document. So that's... Because LibreOffice creates the styles and then names the style of the line. So let's say that we have 100 lines or 100 different colors. LibreOffice will create 100 styles of lines and then before the content of each line will put the code of the style which is, of course, I think the only way because if you change the content of a line the only way is to create a style and identify it. Office of an XML repeats the contents of the style content of a line start of every line. So the number of redundant components is so high that that is really difficult to keep track of everything even for a parser because, you know, you may have one line followed by some strange characters and the combination makes the previous part not easy to understand as the other ones. So my idea is I want to create a very long document making some document like the UNLEAD which is common, which is now public domain and save it as for the FN Office of an XML. So it's a 700 pages document and see what happens.