I agree with the sentiment of the article and the extreme usability of FP. It was easy for anyone to create a website for recipes or knitting designs and to publish it to your corner of the web.
For the longest time, I've felt that an open source project that basically mimicked it would be a great idea. (And for all those widgets, it should be easy to have those operate in asp.net or php... Bring back the page counter!)
For all those static site generators out there (I've written 4), I'd still use a FP clone that edited HTML, rather than markdown, or other.
The closest thing seems to be WordPress, but it lacks one fundamental thing: it's not a desktop app, and requires server config (db and user login).
I think it fair to state that with the demise of FP, (& then flash), the web became too 'professional' as the grandma and 10yo were pushed aside for the semantic web. I still feel strongly that it can all come back with a nice desktop app that worked offline.
The form factor for a casual user is no longer a nice desktop app that works offline - it would now take the form of an online website with a good mobile interface. This is filled by wix, and other sitepage generators like it.
That, and for better or worse, Wordpress. A lot of things that used to have been frontpage sites became Wordpress sites of varying quality and maintenance.
I would say for the better - it's the only thing keeping RSS alive. Although I will also say that having to administer a database is overkill for most use cases.
I loved Frontpage, I used it when I was young (15?) to create an inordinate amount of websites that I ran on one of the IIS servers with Frontpage Extensions, and other doodads, on a computer in my house. That server software came with Windows NT 4.0, which I got in a package of Microsoft software (pretty much everything developer-related that MS sold) I got for free when I went to the MS dev conference in Phoenix (late 90's) because I thought it would be fun. I remember eating quail or some other small game bird at lunch and talking to the real developers who thought it was really cool that I was into computers and showed up at this fancy conference at 15. I don't remember barely any of the presentations, but I do recall seeing one ASP.NET presentation at the time showing an interactive page with a database and everything (whoa! it blew my mind) and I thought that the internet was going to be really cool someday soon after that.
I never became the superman techie that a lot of the folks here became after getting into computers that young, but this article reminded me of those happy days programming with a free version of Visual J++ (why god?!), so thank you.
On another note, as I looked up Visual J++, I read that it was sued out of existence by Sun in 2001, but Microsoft supported the MSJVM for J++ for security updates until 2007! Then, they created J# in order to give those developers a way to transition to .NET technologies, like C#, which was then supported until 2017. Can you imagine any other company supporting old, decrepit technologies like that other than Microsoft? I have a kinda newfound respect for them and their engineering culture of long-term support.
“Can you imagine any other company supporting old, decrepit technologies like that other than Microsoft? I have a kinda newfound respect for them and their engineering culture of long-term support.”
MS seems very willing to support things for a long time. On the one hand that’s a good thing but it doesn’t stop them from constantly cranking out new things and putting them on maintenance quickly. I would much prefer if they stuck to less things and developed and improved them over a long time. Especially in the desktop development area I simply can’t see a coherent strategy. Over the last decade we had WinRT, Winforms, UWP, Silverlight, WPF developed in addition to Win32. I like desktop development but I can’t recommend with a good conscience committing a long term project (in a lot of companies in house software gets used over decades) to one of the MS frameworks because there is a good chance it will soon be obsolete. It doesn’t really help when people say “but it’s maintained” if your chosen framework doesn’t get any new features and there is no upgrade path to the latest framework.
I was looking for a modern "Frontpage-like" editor last year. I wanted something that runs locally, doesn't require a subscription and produces files I can convert to HTML templates for server side rendering.
I bought Pinegrow with their one-time payment after a trial and I'm quite happy with it. Compared to the web-based builders, working with it feels much more responsive.
I use the editor for a few days in a row, but not each month. And thinking about subscriptions for the things that are not central to my workflow just isn't worth it for me even if I could save a few bucks by carefully cancelling/renewing.
This is cool, I'm definitely going to try this out.
I like that they give you a choice of how you want to pay - annual subscription, monthly, or one-time payment. I only have need of something like this occasionally, so I might end up just paying for a month here and there where I need it. More tools should offer this sort of choice.
Back then when I used Frontpage, I couldn't care less about the code quality or the output it generated. It helped me really quickly get a site up and running on the web. Especially personal or family websites. They had wonderful templates for those.
I don't know anything like Frontpage in 2020 that has such a short learning curve but that is as customizable.
I would really pay for something like a modern Frontpage without the bad stuff the author talks about. Despite me being a web developer with a decade plus experience doing websites, I would gladly enjoy working with such a nice tool.
I still remember the thrill of chatting online with my schoolfriends for the first time.
We called each other on our landlines, coordinated which NetMeeting 'directory' to meet on, hung up so we could use dial-up internet, and it worked! It felt magical.
Then MSN Messenger came along and make everything so much easier.
> And then Microsoft bought Skype for $8 billion. And basically killed that too. (I've never done a Skype during Covid. Done plenty Zooms.)
My $0.02:
Skype (for private usage) is still around, and I still sometimes boot it up (to e.g. call my family or recruiters).
Skype for Business is in process of being killed by MS Teams. I only had one call on SfB recently - and it was by far the technically worst experience. Interruptions, people had to turn off webcams to conserve throughput for audio etc.
MS Teams works surprisingly well. I've not had any technical trouble with it during the pandemic, and it's (together with Zoom) the most popular choice.
(I might be biased since I applied to many companies with a MS tech focus)
Great article, and beyond nostalgia, it's also a good lesson about UX and dangers of SaaS.
A tangential question. The article says:
> In the same vein as the Reports, you can see at a glance what links to what. It's a lot like the call chart in a software development IDE.
I read it and I have to ask... which IDE shows nice call charts like that? I've worked with a bunch of those (every Java IDE, C++ in old and new Visual Studio), and there's dearth of any kind of visual representation of anything in general, call graphs in particular. I can sometimes make Visual Studio to begrudgingly show me the static call graph as two trees (one for "who calls me", and other for "who is called by me"), but on a moderately-sized codebase this essentially hangs the IDE.
It's obvious that codebase exploration features would be nice to have. But where are they all hidden? In Enterprise editions of IDEs?
You get reasonably decent call charts from profilers.
Many IDEs include a profiler or support third party profiling, with call charts. Or you can use a standalone profiler which understands the relevant debug info.
An advantage of using profiling is that you get dynamic insight into what may be relatively obscure control flow linkages. The disadvantage, of course, is that it's not complete, if your profiling run didn't include the call, it won't be in the tree.
Author here. I was mostly thinking of National Instruments' LabVIEW and disassembly programs like IDA and Ghidra. Those in particular have actual callgraphs, although LabVIEW is a graphical language anyways so of course it would. You might also see dependency graphs in OOP IDEs, for seeing what inherits what.
I didn't bother to put this detail in the article because this little tangent distracts from the point a bit.
That was a blast from the past. I can’t remember exactly but I think I went Frontpage -> Dreamweaver -> Notepad++ -> Sublime -> VSCode and obviously the other tooling only grew to be more complex (whilst improving sanity). I’m not sure I long for the days of plesk + cPanel etc but I hadn’t thought of all of that nonsense for years. A very different time!
For me it was an Atari Falcon030 with some text editor I can't remember the name of open in one window, and the browser (CAB?) open in another window. When I saved the editor, the browser updated the view without me having to switch windows. Very handy for tweaking.
I was forced to use Frontpage a few times but it was so cumbersome when you were used to programming so I gave up fast and went back to text editors. PSPad and Netbeans lately.
For me it was FrontPage -> Dreamweaver -> Notepad++ -> Spend a month or so dabbling with full IDEs (like netbeans, eclipse, VSCode, etc.) -> Go back to notepad++ -> And recently trying Geany (for cross-platform convenience). My favorite back in the day was dreamweaver (warts and all)...but we're talking circa 2000 ~ 2005. Ah, those were the days! :-)
IMHO what is missing in the analysis is hosting. Sure, a good tool to create HTML/CSS is important, but you also need a server to put that on. The author is just glossing over that fact in their analysis for Weebly/WiX/Squarespace.
These sites used to be scattered all over the web. No need to use other servers. wix, et al came much, much later and was never as free (as in beer and libre).
Those ISPs used to give step by step guides on how to get FP to upload your site.
The magic came when you changed ISPs and you had a different url to upload to... Suddenly your site was in two places (because ISPs then rarely removed old sites)
There was a lot of interesting technology in the 90s that made the bridge between the desktop world and the web world seem-less, Microsoft had a ton of internet integrated into desktop stuff but Mozilla had XPCOM, KDE had Khtml, Adobe had Flash, Java had Applets and a bunch of remote technologies.
I miss that sense of possibility you get from knowing your client side website could do anything.
Of course I can see some of the reasons it went away, security was a constant struggle, maintaining all of the integration points required a lot of developer effort which was being more focused on keeping up with web standards, mobile meant desktop and the web were not the only concerns, etc.
An interesting halfway point was the explosion in widget technologies in the mid-2000-ish time. Often these were web tech based and aimed at similar developers. I think the rise of mobile apps is what killed that one
All the people who used widgets as lazy SEO or injected ads/tracking once they were widespread helped kill widgets off. They had all the same problems as modern browser extensions, but no central authority to remove them.
How insanely clean the UI is. Still not convinced that latest iterations are an improvement. Certainly from a design perspective, the UI looks old here, but not sure about the logical structure of elements. If I look at the tabs of modern Word, I have the impression the categories were chosen at random.
Loved the in-depth analysis, but IMO the article isn't giving enough credit to "Wix, Weebly, Squarespace, etc." at the end. Yes everything pointed out about them is valid, but there is no denying that they are what the class of applications like Frontpage evolved into.
I had limited experience with Frontpage but I remember using Dreamweaver to design my first websites. It had a great tool for automatically rendering the page being designed in preview mode. All back when it was owned by Macromedia before the Adobe acquisition (miss those times!).
For the longest time, I've felt that an open source project that basically mimicked it would be a great idea. (And for all those widgets, it should be easy to have those operate in asp.net or php... Bring back the page counter!)
For all those static site generators out there (I've written 4), I'd still use a FP clone that edited HTML, rather than markdown, or other.
The closest thing seems to be WordPress, but it lacks one fundamental thing: it's not a desktop app, and requires server config (db and user login).
I think it fair to state that with the demise of FP, (& then flash), the web became too 'professional' as the grandma and 10yo were pushed aside for the semantic web. I still feel strongly that it can all come back with a nice desktop app that worked offline.