Flash, Chrome and FireCalc

[marko]

Started receiving a notice on my browser that by the end of 2020 Adobe Flash will no longer be supported on Chrome.

Notice here: https://www.blog.google/products/chrome/saying-goodbye-flash-chrome/

With FireCalc using Flash for it's results page, I'm wondering what the plan is.

No panic, but for us Chromebook users it could become a problem in 12 months.

 

[cathy63]

Are you sure Firecalc is using Flash? It works fine in my Chrome browser that has Flash disabled by default. I typically get the "plug-in blocked" message on a Flash page (e.g. https://get.adobe.com/flashplayer/about/) and I'm not getting that on the Firecalc Results page.

 

[braumeister]

Flash will no longer be supported by Adobe (on any platform) by the end of 2020. It's a horribly bug-ridden product full of security vulnerabilities that should have been replaced on most websites by HTML5. But that replacement is kind of a pain so many sites still use it.

As for FIRECalc, it still works without Flash, just doesn't give you the pretty graph at the end in the Investigate tab (only the numbers).

 

[marko]

Are you sure Firecalc is using Flash? It works fine in my Chrome browser that has Flash disabled by default. I typically get the "plug-in blocked" message on a Flash page (e.g. https://get.adobe.com/flashplayer/about/) and I'm not getting that on the Firecalc Results page.

What I initially see on the result page is a grey/blank page with the Adobe/Flash watermark. It says "click to enable Flash" (or something like that).

Once I do, it works fine until I shut down the laptop and turn it back on.

 

[braumeister]

Yes, but by allowing your browser to use Flash like that you're exposing yourself to security risks. You might be fine, but you might not. Just something to be aware of.

 

[marko]

Yes, but by allowing your browser to use Flash like that you're exposing yourself to security risks. You might be fine, but you might not. Just something to be aware of.

Thanks but yes, I've known that but over the years we had few, if any, alternatives.

I'm not going to miss Flash; just asking if the admins of FireCalc have a plan to replace it.

 

[USGrant1962]

Thanks but yes, I've known that but over the years we had few, if any, alternatives.

I'm not going to miss Flash; just asking if the admins of FireCalc have a plan to replace it.

Unfortunately, it seem unlikely. The current owners have not addressed a few true bugs that have been demonstrated here over the years, and the poverty lines on the variable spending charts are from 2006 (a two minute fix). Every once in a while someone on the forum volunteers to do some update work to no apparent effect.

 

[razztazz]

Unfortunately, it seem unlikely. The current owners have not addressed a few true bugs that have been demonstrated here over the years, and the poverty lines on the variable spending charts are from 2006 (a two minute fix). Every once in a while someone on the forum volunteers to do some update work to no apparent effect.

Holy cow. I didn't know things were quite that bad. I'll be looking for a good back-up / new primary calculator

 

[ERD50]

...

As for FIRECalc, it still works without Flash, just doesn't give you the pretty graph at the end in the Investigate tab (only the numbers).

But some of the 'Investigate' options use a graph for results, with no text, portfolio survival versus AA is one. So you get zero useful output.

The last time someone posted that chart, I took a screen shot.

Sure would be nice to see updates and bug fixes to FIRECalc, it gets used a lot around here.

-ERD50

 

[Andy R]

Thanks for the heads up, we will look into redoing FIRECalc in the next year. IMO we should be taking a mobile first approach to the redesign and build it using technology so it can also work as an app. I also believe we should evolve FIRECalc so it remembers your data in a database vs the URL as the URL method is not secure from prying eyes.

FIRECalc is a worthwhile resource & it deserves some TLC. The reason it hasn't gotten much attention is that other opportunities with better ROI have been taking precedence. It's not like we are a big brokerage that can justify some hefty development expense to provide a tool like this for free (the ad revenue is miniscule).

Now, with a deadline, it seems like the right time to take some steps forward. Do we have any UI/UX gurus who want to help? How about any software architects or React/Laravel developers?

@USGrant1962 - Any chance you can get me a list of the known bugs. I will queue those up for being fixed. Background data updates we can get done fast. Other code related stuff we would need to evaluate.

 

[USGrant1962]

Thanks for the heads up, we will look into redoing FIRECalc in the next year. IMO we should be taking a mobile first approach to the redesign and build it using technology so it can also work as an app. I also believe we should evolve FIRECalc so it remembers your data in a database vs the URL as the URL method is not secure from prying eyes.

FIRECalc is a worthwhile resource & it deserves some TLC. The reason it hasn't gotten much attention is that other opportunities with better ROI have been taking precedence. It's not like we are a big brokerage that can justify some hefty development expense to provide a tool like this for free (the ad revenue is miniscule).

Now, with a deadline, it seems like the right time to take some steps forward. Do we have any UI/UX gurus who want to help? How about any software architects or React/Laravel developers?

@USGrant1962 - Any chance you can get me a list of the known bugs. I will queue those up for being fixed. Background data updates we can get done fast. Other code related stuff we would need to evaluate.

Andy, this is very exciting news

Bugs:

1. Past date entry - is a sticky on the forum, but still pops up regularly. Needs a data entry check/warning.

2. 16% Problem. Entering a small portfolio with large pension tumbles to an arbitrary 16% withdrawal rate.

http://www.early-retirement.org/forums/f36/firecalc-question-94602.html
Firecalc spending analysis - Early Retirement & Financial Independence Community

3. Min ending portfolio equals starting portfolio.

lowest portfolio = starting portfolio - Early Retirement & Financial Independence Community
Lowest portfolio balance - 100% success - Early Retirement & Financial Independence Community

4. Bernicke bug, second graph is wrong for future retirements (or perhaps just confusing?)

FIRECalc Spending for Future Retirement Question - Early Retirement & Financial Independence Community

5. "What if" text error.

FIRECalc glitch? - Early Retirement & Financial Independence Community

6. SS input bug (might not be real, could be a past date issue).

May have found bug when entering Soc Sec - Early Retirement & Financial Independence Community

Minor updates I would like to see:

1. Reset the default to $40,000 spend, $1,000,000 portfolio
2. Update the poverty levels for the percent portfolio spending graphs
3. Change the default expense ratio to say 0.10% to better reflect the current market
4. Add the lowest spending amount to the text on the variable spending output.

A nice moderate update would be to rework/update the "Mixed portfolio" categories and add foreign stocks.

I'm sure the members here can generate an extensive wish list of updates, some easy and many hard.

Good luck and thank you for hosting FIRECalc.

 

[razztazz]

A nice moderate update would be to rework/update the "Mixed portfolio" categories and add foreign stocks.

Along these lines, one option I always liked to see in a calculator of this type is the option to NOT re-balance. Just set and forget.

 

[marko]

I would only add the odd results you get if you enter taking SS in a year earlier than today's date.

If you were to enter that you started taking SS two years ago, the results get all skewed.

But I'll be happy just to get it working without Flash!

 

[ERD50]

....

Minor updates I would like to see:

1. Reset the default to $40,000 spend, $1,000,000 portfolio
2. Update the poverty levels for the percent portfolio spending graphs
3. Change the default expense ratio to say 0.10% to better reflect the current market
4. Add the lowest spending amount to the text on the variable spending output.

A nice moderate update would be to rework/update the "Mixed portfolio" categories and add foreign stocks.

I'm sure the members here can generate an extensive wish list of updates, some easy and many hard.

Good luck and thank you for hosting FIRECalc.

Good consolidated bug list. I would hope the admins here would have been keeping one.

I bolded the 'wish list' that to me is an obvious improvement (I manually do it almost every time, as the results convert to % with just a decimal shift). This must be a very, very simple update to the current FIRECalc, no programmatic/logic changes, just a variable set/definition somewhere in the code. It would be nice to get this one as a New Year's gift. While I don't have credentials as a programmer, I have enough experience to know this is a very minor and uncomplicated change.

Along these lines, one option I always liked to see in a calculator of this type is the option to NOT re-balance. Just set and forget.

Yes, that is handy, just to see if it has any big effect or not (usually bot, sometimes negative). However, I'd still want the option to make the withdrawal such that it helps to get back to target AA, if it has got out of balance. And ladies and gentlemen, there's our first 'feature creep'.

I'd also like to see a complete revamp of the results graph. The squiggly lines jumbled together just don't convey enough info. I'll PM Andy about this.

And while I have no credentials in UI/UX, I (along with most of us) certainly know bad UI/UX when we see it, and given a clean slate, I can also envision how things should be for ease of use/clarity. So I'd be willing to help in that regard, and as an alpha/beta tester, but...

To be frank, we've been down this road before. You were working with us to do a revamp the interface some years ago, as a prelude ti updating the code, and the project just got dropped after a fair amount of input from us. So excuse me if I'm a bit skeptical that this will get enough attention and priority that I should bother with it.

Andy, I'll PM you about some other things that I can't share with the others here.

-ERD50

 

[tb001]

A few things I would love to see:

Starting dates for the cycles that fail. A different calculator has this and it’s remarkably informative.

More lines to adjust spend or maybe a start and stop date, so if you have a mortgage you don’t have to use two lines.

Making it easier to find manual spend option might help on the $ side. I had used it for easily a year before I stumbled upon a thread here and went looking for it. Or maybe charging for the option to save your data or specific scenarios?

IMO, it’s one of the better tools available and for many would be worth a small fee for added features.

 

[Andy R]

Good consolidated bug list. I would hope the admins here would have been keeping one.

Just an FYI - We don't have an Admin assigned to this. The team here has been focused on managing this community, they were not asked to organize any lists of bugs or feature requests.

To be frank, we've been down this road before. You were working with us to do a revamp the interface some years ago, as a prelude ti updating the code, and the project just got dropped after a fair amount of input from us. So excuse me if I'm a bit skeptical that this will get enough attention and priority that I should bother with it.

Well, on the previous push, we did actually make progress on backend code. It was nothing anyone could see but I had a developer refactor a bunch of the code so we could work on it easier. It's been some time since then and we likely need to do this again and get FIRECalc onto a more modern development framework.

What I'm hearing is that people want more features and that there might be a market for offering them as an upgrade. What's been taking my attention away in recent years is building SaaS tools for RVers since we run a bunch of RV websites. We are able to market the tools to our readers and slowly start to diversify our income from primarily ad supported to also offering SaaS services.

What I learned is that the right business model gets the love/attention. For example, a website might make miniscule money and thus can't afford to invest in itself. Alternatively, if you can find a business model that generates income then you have funds to reinvest into your product.

Based on the feedback above, it sounds like there could be demand for free FIRECalc and Pro version. The Pro version would have some type of annual fee. We would then put some of the advanced features in the Pro version in hopes people would see the value and upgrade.

Finding a business model that allows FIRECalc to support itself (ie. earns money so we can then spend that money making it better) would be ideal. I'm not talking huge numbers or anything but if FIRECalc could make some money, then we could spend that money back on improvements to FIRECalc.

Andy, this is very exciting news

Thanks for that long list and reminder to go check the FIRECalc forum.

 

[MuirWannabe]

Firecalc may have a few bugs today. But nothing major imo, and nothing impacting its fundamentals for helping aspiring FIRE’s determine the viability of their situation.

I wasn’t around when the current Firecalc was built and cleaned up. But having been in the IT business, I’m well aware how long it can take to get a new application up to snuff and reliable. And trustworthiness of a tool like Firecalc is critical. People are basing life changing decisions on its results being trustworthy.

I’m not objecting to a revamp/rewrite of Firecalc. In fact, it’s an exciting prospect. But the QA and testing required to get a new version to be as reliable as the current version with its vast number of executions over the years is intimidating. Please bake this into the plans and then add some extra. QA is like bacon. You can’t have enough. Also, please build walls around the current Firecalc code, so that the original always exists. Or at least until a new version overwhelms it in terms of confidence.

Thanks for your consideration.

 

[USGrant1962]

...
What I'm hearing is that people want more features and that there might be a market for offering them as an upgrade.

....

I was thinking of suggesting that for my wishlist. Things included in a FIRECalcPro subscription might be:

• Print/PDF reports
• Save cases (beyond just saving the url), compile comparison tables/charts for multiple cases/scenarios
• More "changes" in addition to the current subscriber spending model inputs (e.g., changes in asset allocation, more "other income" entries, more "lump sum" portfolio entries)
• Renovate/update the "mixed portfolio" section and move it to Pro
• Upgrades to the output, like choosing graph scales, click to identify the starting year of a line, etc.
• Maybe a more robust Monte Carlo / random performance function

So I guess I'm suggesting that any and all upgrades you apply could become part of the Pro package, and perhaps some of the more specialized current features as well. The side benefit of the latter is it might make the standard version even cleaner and easier to use for new folks.

Separately, members here (and ER types in general) tend to be really [-]cheap[/-] frugal, so pricing of the Pro version may be an issue.

 

[tb001]

Lots of good suggestions. If you’re considering adding features, the fidelity RIP tool is also great. SS and tax modeling would be interesting expanded features, but lots of work I know.

 

[Andy R]

I was thinking of suggesting that for my wishlist. Things included in a FIRECalcPro subscription might be:

• Save cases (beyond just saving the url), compile comparison tables/charts for multiple cases/scenarios

Separately, members here (and ER types in general) tend to be really [-]cheap[/-] frugal, so pricing of the Pro version may be an issue.

Keeping values in a URL string is terrible. Even more so now that net neutrality laws were repealed. There is just too much info that can be observed via the URL even over SSL.

What I was thinking was to put a lot of focus on security & privacy. I think appreciate that and might be willing to pay for that these days. Basically we could architect the system so you could store your data or multiple iterations of your data (ie. scenarios) in the database. Since the focus would be on security, we would encrypt information in the database so that even if hacked, the data would be worthless. Basically one of the selling points would be that we are not tracking you, that your data is private and safe.

As for frugal [-]cheap[/-] folks, we can try to find the right balance. It's really hard to build a sustainable project when free.

 

[ERD50]

... Also, please build walls around the current Firecalc code, so that the original always exists. Or at least until a new version overwhelms it in terms of confidence.

Thanks for your consideration.

+1 on keeping the current version intact until there is confidence in any new version.

As far as testing to gain confidence, here's a suggestion: Allow the option of accepting inputs from a plain text/csv file. This would let testers create a suite of standard inputs (or ad hoc as desired), and easily run them to test the output against an expected result. And if the output could be dumped to a text/csv file, an entire suite of tests could be automatically validated. That's a heck of a lot easier and error free than typing stuff into fields on a web page. And easy to share compared to saving a bunch of profiles/cases under a user name.

.... As for frugal [-]cheap[/-] folks, we can try to find the right balance. It's really hard to build a sustainable project when free.

Absolutely. And if you were to tell us you just can't support this, because there just is no money in it for you, I'd respect this. It's a business, not a charity.

I do hope you can find ways to profit from it. I know next to nothing about making money from services like this, so I have nothing to offer there. But even a cheapskate like me would be willing to pay something ($10/year is a nice round, picked out of the air number) for something above a very basic set of features. I think if you limit a 'pro' version to only very advanced features, the audience would also be very small. Maybe you'd need to go very advanced, and attract financial planners/pros - a much smaller audience, but maybe willing to pay much bigger bucks? But again, I know nothing in this area.

What about an open source approach? Again, I have no idea how these are managed, but there seem to be a lot of interested coders out there, and I've learned that there are some free libraries out there that seem to be easy to use that provide some very fancy and useful graphing options.

-ERD50

 

[Andy R]

What about an open source approach? Again, I have no idea how these are managed, but there seem to be a lot of interested coders out there, and I've learned that there are some free libraries out there that seem to be easy to use that provide some very fancy and useful graphing options.

We would likely build it using lots of open source tools but probably not likely to open source the code itself for a few reasons.

1) It would likely undermine a Pro service as people could just setup and offer it for free.

2) Being open source means everyone can see your code including any potential exploits. Of course, this is a good thing if you have an active development community. Building such community is pretty tough and most projects get abandoned.