I built my current house recently, and our builder used a prefab factory in the region for some of the construction. I would _never_ do it again. The process has all the pitfalls of the waterfall software development model, but at the end you have to live in the failed project!
It took months for the factory to do their pre-build engineering, and that's after they started the process using plans from my architect. The companies building these modules tend to sell mostly from their own catalog of floorplans and modules, so they don't have a strong incentive to hire skilled architects for the relatively small amount of custom work that comes in (I fell into this category). I'm led to believe that they hire draftsmen instead of architects.
They built one of the modules a foot too long, so it hangs over the edge of the foundation. They also messed up the roof line, so some guttering and geometry had to be changed for the garage that was stick-built onsite. They did all of the drywall wraps around the windows incorrectly. Every single door was hung incorrectly, some with chunks taken out from bad routing. The vinyl plank flooring has bubbles because they didn't float it correctly. They omitted electrical features that were on their own drawings. They ordered a few wrong windows, so they charged me anyway and just laid them on the floor when they shipped the module.
It'll all be fine at the end of the day, but the parts that were built onsite are all so much better. No cheaper, mind you, but faster and with much less drama. Maybe there are some parts of the industry that do a lot better, but I kind of wish I'd listened to my dad when he said "the pioneers take the arrows".
> […] so they don't have a strong incentive to hire skilled architects for the relatively small amount of custom work that comes in (I fell into this category). I'm led to believe that they hire draftsmen instead of architects.
If you're going to go with a vendor with a 'catalogue' of prefab plans, then I would think you'd want to choose from their pre-designed plans.
There's a middle-ground between completely prefab catalogue plans, and stuff that's built by on-site carpenters (which aren't a bad option if you're doing custom) doing 'stick built': and that's to hire a company that specifically does custom prefab frame walls:
Yup. The theme is "why isn't prefab assembly line production not taking over home building", and the (current) top comment is: prefab didn't work for my custom design.
While architects can be talented and make a great design on so many levels, can they really fully optimize space for usability and features to a deep level? I just think of RV interior design, which amazes me sometimes with the ingenious use of space that is clearly the result of dozens of rounds of iterative design over the year.
Not that a house should be that crazy (aside from the tiny house movement) with maximizing space, but if there were industrial production, then you could get to deep optimization of the floor design.
What always strikes me with (american suburban) houses, is that they are very inefficient at vertical storage of things like clothes, dishes, and a ton of other housewares that you need on a daily basis (so you don't dump them in the basement). I walk through 3000 and 4000 sqft houses and they are only marginally better than a well designed 2000 sqft house in usability.
You ever notice why all the people satisfied with their custom home are either obscenely wealthy or the kind of people who buy a front end loader just because it's a good deal even if it needs work? There are two ways to get a custom home done right.
1. Be stupidly rich and don't give a crap about how much you spend
Absolutely correct. In fact to expand on this slightly... I've had a lot of clients wanting to find land to build...what they don't understand is how few lenders are willing to lend on just land / single home building projects (at least in my market). Then you add in the well and septic where so many just waive their hands "no big deal right?" Um - it sure is a big deal if you are the "lucky" party who doesn't hit water for > 1000 feet. I've got a lot of prospective buyer / builders who show up with tech money and end up very surprised at how quickly you can eat through a million dollars cash when you start from bare land.
This. We live in farm country and bare land loans have gone from 20% down to 25% down in the last couple of years. You receive loans in chunks for building a custom home as the builder completes milestones. There was a "restoration" loan that came with certified builders and a government mandated architect that signed off on payment to the builder as milestones were reached. I am not sure if this type of loan is available any longer. In my opinion it is very difficult to purchase a mini farm (less than 100 acres) and make a go of it in the south. Some of our friends are building "barn" homes. Buy the land cheaply 5-10 acres. Have a metal barn built for less than $60k and then convert the interior to a home. This is very popular currently in Carroll county GA (feeder to Atlanta for jobs).
Although I think throwaway is correct that that isn't what they meant, people do indeed lease land in the United States. It's more common in commercial real estate, but there are also (predatory, in my opinion) land leases for "manufactured" (i.e. mobile) homes where the resident buys the home but doesn't own the land under it.
Yes, throwaway understood my meaning... and you are right too that there are circumstances where people own a home but not the land underneath. Apart from the example you gave, I'm also vaguely aware of cases where tribes own reservation land where homes have been built by non-tribe members who technically have a lease to the land.
And actually you bring up another wrinkle, which is if you don't own the land underneath, there are plenty of lenders who also won't lend to you to buy a manufactured home, because they consider it to be personal property rather than a permanent dwelling (which they could more easily foreclose and sell)
Isnt that the point of the article - the "bone structure" is supposedly built so that unskilled labour (ie you and three mates) can erect the building yourself?
I have to admit that having a prefab building that has one side a foot longer than the other side not sound like a condemnation of the concept but the factory that built it. I presume one would have recourse under contract for such a foul up?
> I presume one would have recourse under contract for such a foul up?
Possibly, but delays in construction are what kill you. You can't really FedEx overnight new structural steel framing components, so now you're blocked until they can get
It sort of sounds like perhaps if you are going pre-fab you shouldn't expect to be able to do any modifications or fit their pieces to your own plans? Instead just find a site that is suitable and plop down their exact premade floorplan?
I have no evidence, but I suspect that's likely to get the best outcome, as far as the factory's quality level is concerned. They're pretty heavily optimized for a particular set of processes, and anything that departs from that happy path seems likely to increase risk.
Yes, this story reminds me of some clients I’ve had that always want to customise off the shelf software to do what they want when it would be easier to build it from scratch.
After it was delivered (and it comes delivered with sheetrock walls, electrical, and subfloor), my coworker noticed that the the factory forgot a step: the plumbing. The builder had to rip out lots of the sheetrock and install plumbing.
Lots of mistakes occur while building a house. The difference is that when the workers are already in the field, it is generally quite inexpensive to identify and fix any problems. Not so for the prefab stuff.
I think these problems are unique to your type of home maybe.
I have been to a Bone Structure presentation. They fabricate all parts and interfaces using computer operated cutters with extreme precision from CAD plans. Note that parts are made of metal, not wood.
The structure is then aligns perfectly and all connection holes align perfectly as well.
The building is then assembled using just an electric screw driver, on sight. They claimed that almost anyone could do it, like IKEA. But they still require skilled technicians. But the point was that it isn’t a rocket science and plans are designed in such a way that you can’t really mess it up too much.
It may be precisely fabricated exactly as specified in the CAD plans. That doesn't mean the CAD isn't riddled with errors.
I've worked in traditional construction and the plans are never accurate with hundreds of omissions/errors/interferences. Those can be corrected in real time when something is field built, but are nearly impossible when the house ships wrong but nearly finished.
I can see it working with cookie cutter housing - first ~5 will be garbage as they find bugs then slap a thousand identical copies up cheaply. For custom it's a mess.
Getting on site to realise X is completely wrong is so standard most architects spend probably 20%+ of their time talking to builders. This level of QA never happens in a factory.
The buildings also look horrendous by default, artificial and cheap looking with a complete lack of character. I’m all for prefab in theory but it’s like building a whole piece of software without testing any of the interfaces fit together or that say logging and error handling are plumbed in without testing them. It’s clearly a recipe for disaster even with the best organised prefabricator.
We had a frank conversation with the builder about costs, as well as what was and wasn't worth fighting over. Some items were able to be repaired without impacting the budget by subcontractors that were already coming in to do other things. Other things we didn't care enough about to make a fuss over (like the roof line not matching up right on the back side of the house, where no one will ever notice). Some things we asked her to fix at her expense.
We were very aware that big projects, be they software or home construction, don't ever go perfectly, so we made a conscious effort to stay calm and solution-oriented through the entire multi-year process. We came in very slightly under budget, and things that need to get addressed are getting addressed, whereas a lawsuit would have turned things adversarial very quickly.
>We were very aware that big projects, be they software or home construction, don't ever go perfectly
Most large projects have some central features/processes that are by their very nature both very expensive and very complicated.
The entire prefab industry is built around minimising those things - but then something ELSE becomes the most expensive/complicated thing. You then have a situation where if you're seen as a bottleneck, holding up the expensive/complicated thing, then the total cost of the project will balloon.
Many prefab companies know this and are happy to sell modules with little QA work done, because it's unlikely that there will be serious repercussions. Once you accept the work then it'll be too hard for your legal team to fight for (or at least, not financially worth it).
I'm not saying that prefab companies are ethically bankrupt, but it IS an environment that promotes this method of doing business. I've been in construction for mining/oil&gas companies, on both sides of this situation, quite a few times. The amount of shoddy work you're willing to accept is quite terrifying once the cost for re-work gets into nine and ten digits.
I imagine it's because he would be largely up against a company and litigators skilled at meeting the letter of the contract if not the spirit. "Winning" would likely see him get a minimum level of corrections done to the house, which could be accomplished for less than the cost of a lawsuit.
According to my neighbor, who is a lawyer and likes to drink, this is pretty common. He is on a lot of cases where the lawsuit makes him (the lawyer) more money than either party would have saved or lost had they just settled up front.
The one I remember had to do w/ $1900 worth of lost wedding photos cuz a server died or got deleted. Cost more to twist the knife in court than to just settle.
Probably depends on your local jurisdiction... I am not a lawyer but my own experiences tell me that any kind of civil case below about $50K, the courts would really you rather not bother them with this, and you are going to spend at least 20% of that and several years before you may or may not see a dime.
Oh and the 15k are only for legal costs. Don't forget rent for 3 years on top of that. And all the money that you'll still have to pay for the construction. And stress (medical bills are not cheap).
It may not have been the customisation, more the processes of pre-fab in general and what the production process is. I didn't build my house but it's a recent pre-fab build, straight from the catalogue (a Swedish company). The amount of shoddy workmanship and shortcuts taken is surprising and I'm slowly correcting replacing a lot of the door & window surrounds, some floor supports, broken wiring, badly aligned panels, etc...
The overwhelming downside of not needing professional builders to build a house is not having professional builders build a house. Half-assed amateurs aren't going to give you the quality you expect for your home.
I can imagine that bespoke is not a great fit for prefab. My impression of prefab construction is that it's mostly used when lots of identical buildings need to be built. Then you can afford to have everything tested in advance to make sure it all fits together. If you're building something unique, don't you basically miss out on all of the advantages of prefab, while keeping its disadvantages?
Somebody in my organization got very interested in Dgraph in its 0.7 or 0.8 days. The version was marked as "production ready", but it was an absolute trainwreck.
We were modelling individuals and contacts between them, and the cluster would constantly break with dataset sizes that should have been easily managed. There was clearly something wrong in the storage engine, because we saw insane disk space usage. Dgraph consumed 10s of TB for something that should have taken < 100 GB.
We were one of the largest installations at the time, and were working with the core development team, but they were never able to resolve the issues.
We eventually had to tell management that there was no way we'd be able to operate the thing given its disk space consumption rate, so we had to delay project delivery to rip out Dgraph and replace it with Postgres.
Surely it's better today, but I'll never use it again by choice.
Dgraph is built for performance and with one of our app, we faced similar challenges. After reading some documentation and watching some of their videos, I got to know they compromise space against performance when we have lots of index.
We reduced index from 35 to 8 and the db size got drastically low.
I believe you should investigate that as well to check if it's wrong in your database architecture design.
For your gap of <100 GB against 10,000 GB. I assume, you probably created lots of index.
Just create good database design and reduce index, you will have low size high performance app.
Just my two cents, but if I was your coworker and you told me the situation, I can virtually guarantee that my annoyance would instantly transmute into some degree of sympathy and tolerance. It goes a long way to know that you're aware of the problem and feeling some stress yourself. Suddenly it isn't something you're inflicting on me because you're oblivious.
When I was a kid, I got to take some Logo programming classes at Purdue. I can't imagine how much more intuitive and visceral it would have been if we'd had Minecraft instead of blank white screens.
I happen to play Minecraft now, and I just finished writing a 3D Perlin noise generator in Lua so that my turtles can build structures with some randomization. I'll be adding code to pixellate my smooth, mathematical paths into minecraft blocks using Bresenham's line algorithm. I've also written programs to do highly efficient mining with minimal effort on my part. I've seen programs written by other people that control a fleet of turtles from a central computer, and programs that run complex farming operations. There's so much breadth of possibility with something like Minecraft and ComputerCraft.
If Minecraft is like playing with Legos online with your friends, ComputerCraft is like Mindstorms. I envy the kids who start out on this stuff.
